Novel 18F-Labeled Radioligands for Positron Emission Tomography Imaging of Myelination in the Central Nervous System

Article abstract of DOI:10.1021/acs.jmedchem.8b01354

Myelin is the protective sheath that surrounds nerves in vertebrates to protect axons, which thereby facilitates impulse conduction. Damage to myelin is associated with many neurodegenerative diseases such as multiple sclerosis and also includes spinal co

Full text of DOI:10.1021/acs.jmedchem.8b01354

Article
pubs.acs.org/jmc
Cite This: J. Med. Chem. XXXX, XXX, XXX−XXX
18
Novel F‑Labeled Radioligands for Positron Emission Tomography
Imaging of Myelination in the Central Nervous System
Anand Dev Tiwari,† Junqing Zhu,† Jingqiang You,§ Brendan Eck,‡ Jinle Zhu,† Xu Wang,⊥ Xizhen Wang,∥
Bin Wang,⊥ Jerry Silver,§ David Wilson,‡ Chunying Wu,*,† and Yanming Wang*,†,⊥
†Department of Radiology, ‡Biomedical Engineering, and §Neuroscience, Case Western Reserve University, Cleveland, Ohio 44106,
United States
∥Department of Radiology, Weifang Medical University, Weifang, Shandong 261053, China
⊥Department of Radiology, Binzhou Medical University, Binzhou, Shandong 256603, China
S
* Supporting Information
ABSTRACT: Myelin is the protective sheath that surrounds
nerves in vertebrates to protect axons, which thereby facilitates
impulse conduction. Damage to myelin is associated with many
neurodegenerative diseases such as multiple sclerosis and also
includes spinal cord injury (SCI). The small size of the spinal
cord poses formidable challenges to in vivo monitoring of
myelination, which we investigated via conducting a structure−
activity relationship study to determine the optimum positron-
emitting agent to use for imaging myelin using positron
emission tomography (PET). From these studies, [18F]PENDAS was identified as the lead agent to use in conjunction with
PET imaging to delineate the integrity of spinal cord myelin. A subsequent in vivo PET imaging study of [18F]PENDAS in rats
with SCI showed promising pharmacokinetic results that justify further development of imaging markers for diagnosing myelin-
related diseases. Additionally, [18F]PENDAS could be valuable in determining the efficacy of therapies that are currently under
development.
approaches, which may be more effective in reducing long-term
disability.8
INTRODUCTION
■
Spinal cord injury (SCI) is one of the most catastrophic traumas
to the body that affects 250 000−500 000 people worldwide
annually. A traumatic blow to the spinal cord typically crushes
the nerve fibers, which causes apoptosis of oligodendrocytes and
subsequent loss of myelin.1 Diagnosis, prognosis, and efficacy
evaluation of SCI treatments require imaging tools that permit
the detection and quantification of myelin status that correlates
with the degree of physical disability.
In the clinic, SCI is primarily assessed through magnetic
resonance imaging (MRI), which is a nonspecific imaging
modality that only detects overall lesion load in the spinal cord;
however, specific myelin changes from onset to late stage have
important implications for diagnosis and prognosis. This is
particularly true in SCI where myelopathy becomes the primary
characteristic of clinical presentation with extensive abnormal-
ities in the spinal cord. It is thus critical to detect and quantify the
extent of demyelination and remyelination and elucidate their
relationship with physical disability.
For accurate diagnosis and evaluation of myelin repair
therapies, it is critical to develop a molecular imaging tool that
allows for direct detection and quantification of myelin changes
in the spinal cord. Because MRI signals only reflect changes in
tissue water content that do not distinguish myelin damage from
inflammation or edema, conventional T2-weighted MRI lacks
pathological specificity. Other types of more technically
sophisticated MR imaging methods such as fluid-attenuated
inversion recovery and diffusion tensor MR imaging are also
nonspecific in detecting changes in myelin. Recently, an MRI
technique using magnetization transfer ratio has also been
explored to monitor myelin change.9 However, this technique
still does not provide sensitive and quantitative measures of
myelin but can only be used as an indirect measure of myelin
integrity.10−12
To address this challenge of myelin imaging in the spinal cord,
we set out to employ positron emission tomography (PET)
imaging, a widely used clinical imaging modality that has
extremely high sensitivity as well as quantitation capability. To
optimize the advantages of PET, we synthesized a series of small
molecular probes (SMPs) with myelin-binding capability. Some
At present, most of the available drugs for treating SCI
decrease inflammation through modulation of the immune
system. New myelin repair therapies are currently under
development that are designed to promote remyelination.2,3
A
new approach to SCI therapy is that myelin repair can be done
Received: August 30, 2018
via transplanting myelinated stem cells4−6 and/or antibody7
© XXXX American Chemical Society
A
DOI: 10.1021/acs.jmedchem.8b01354
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 1. Structures of myelin-imaging agents that have been developed based on stilbene.
of the SMPs can be readily radiolabeled with positron emitters
such as C-11 or F-18 that enable us to apply PET to detect and
quantify myelin integrity in the central nervous system (CNS).13
Subsequent imaging studies showed that these SMPs primarily
labeled with C-11 bind with high affinity and specificity to
myelin in vivo and can be used as an imaging marker to
efficaciously evaluate myelin repair therapy in the spinal cord.
In addition to developing a series of myelin-imaging agents for
PET imaging, we have also developed agents that can be used in
conjunction with single-photon emission computed tomog-
raphy, MRI, and optical imaging (Figure 1).14−25 Among these
probes, methyl diamino stilbene (MeDAS), case imaging
compound (CIC), myelin-binding agent (BMB), (E)-4-(4-
amino-3-(2-fluoroethoxy)styryl)-N-methylaniline (FC2Me-
DAS), and triazole fluorinated diamino stilbene (TAFDAS)
were radiolabeled and successfully used with PET imaging with a
variety of animal models.14−19,26,27 Most of these imaging agents
share a common diamino stilbene (DAS) pharmacophore,
which facilitates specific interaction with myelin sheets. The
planar structure of DAS makes it possible to bind to the unique
β-sheet assembly present in myelin basic protein (MBP).24 The
exact binding mechanism is still unknown due to the complex
composition of myelin.
Our previous studies focused on C-11 labeled radiotracers
such as BMB, CIC, and MeDAS for in vivo PET imaging of
myelin to take advantage of its relatively short half-life and
predictable radiostability after radiolabeling.14,16,18 Completion
of these studies provided proof of principle that SMPs can be
developed for PET imaging of myelin and used for efficacy
evaluation of remyelination therapies. Thus far, we have
identified [11C]MeDAS as a lead PET radiotracer to monitor
changes of myelin in the spinal cord. To extend the availability of
these agents to a wider audience of users, we developed a series
of MeDAS analogues radiolabeled with fluorine-18, which has a
significantly longer half-life and therefore would permit remote
distribution to imaging facilities without onsite cyclotrons. We
first developed a series of [18FMeDAS analogues such as
FC2MeDAS, as shown in [18F]MeDAS analogues such as
FC2MeDAS, as shown in Figure 1.27
To further optimize the in vivo properties and pharmacoki-
netics in the spinal cord, we designed a novel series of fluorinated
MeDAS analogues by direct alkylation of one of the amino
groups of DAS. After synthesis and radiolabeling, the binding
properties and pharmacokinetics of these fluorinated stilbene
analogues with different terminal functional groups were
evaluated through systematic structure−activity relationship
(SAR) studies, which enabled us to identify a novel radiotracer
that is ideally suited to ascertain myelin integrity via clinical PET
imaging.
RESULTS
■
Chemistry. For SAR studies, a series of stilbene derivatives
was designed and synthesized, as shown in Table 1. Based on
[11C]MeDAS, a C-11 labeled radioligand,18,28 we introduced N-
1-fluoropentyl amino groups to a series of stilbene analogues
containing −NH2, −NO2, −NHCH3, −OCH3, −SCH3, and
−CH3 functional groups. In addition, a fluoropentyl chain was
selected based on SAR studies, as previously described in J. Med.
Chem. 2016, 59, 3705−3718. In that study, we tested a series of
compounds with various lengths of fluoroalkyl groups. SAR
studies suggested that longer fluoroalkly groups such as
fluorethoxyethoxy showed optimal lipophilicity and radio-
chemical stability. Fluoropentyl has length similar to fluoroe-
thoxyethoxy and is widely used in other fluorinated agents for
PET imaging, such as [18F]-2-(5-fluoro-pentyl)-2-methyl
malonic acid for cancer imaging,29 18F-(fluoropentyl)-
triphenylphosphonium salt for myocardial imaging,30 and
[18F]3-(4,5-dihydrooxazol-2-yl)phenyl (5-fluoropentyl)-
carbamate for neuroimaging.31 We thus selected fluoropentyl
to introduce F-18 to myelin-imaging derivatives.
B
DOI: 10.1021/acs.jmedchem.8b01354
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
nitrobenzaldehyde (a) to produce the precursor (9a) through
the Horner−Wadsworth−Emmons reaction in 40% yield. This
precursor (9a) was used for the radiosynthesis of 18F-labeled 5a
for in vivo microPET/computerized tomography (CT) studies.
We then systematically evaluated these newly synthesized
fluorinated stilbene derivatives (5a−f) in vitro as well as in vivo
and ex vivo to determine their properties of binding to myelin.
The UV absorption and excitation/emission spectra of all of the
final products (∼1 mM) were first acquired in acetonitrile
(MeCN), as shown in Figures 2 and 3, respectively. The
absorption wavelengths ranging from 340 to 425 nm were
observed for the final compounds (5a−f). The excitation
wavelengths ranging from 350 to 410 nm and emission
wavelengths ranging from 415 to 575 nm were determined for
all the test compounds (5a−f). Such wavelengths render these
compounds highly fluorescent and suitable for fluorescent tissue
staining to evaluate the myelin-binding properties in vitro or ex
vivo.
Table 1. List of the Final Compounds (5a−f) with Their
Calculated log P (clog P)
In Vitro Fluorescent Tissue Staining. We first conducted
in vitro fluorescent tissue staining to evaluate the myelin-binding
properties of the newly synthesized compounds. Figure 4A
illustrates that compounds 5a−f stained the myelinated regions,
including the corpus callosum and striatum, where the
fluorescent intensity correlated with the myelin distribution
pattern in the brain. To quantitatively compare myelin-binding
potentials, the test compounds were screened in vitro in mouse
brain tissue sections at the same concentration (1 mM) for 25
min, and tissue sections were imaged under a microscope (Leica
DM4000B) at the same exposure time. After staining, we
defined two standard regions of interest (ROIs), one from the
genu corpus callosum (gcc) that is myelin-rich and the other
from the subcortical gray matter (cortex) that is myelin-
deficient. These differences are illustrated in Figure 4A, from
which the fluorescent intensity was calculated to determine the
fluorescence intensity ratios (FIRs) between the two ROIs
(Figure 4B). As a result, compounds 5a and 5c show the highest
FIR values >1.2 among all of the newly synthesized compounds
(Table 1). This SAR study suggests that compounds 5a and 5c
exhibit high in vitro binding specificity, justifying them as lead
candidates for subsequent ex vivo evaluation.
Ex Vivo Imaging of the Spinal Cord and Brain. To
determine if the lead compounds bind to myelin of the spinal
cord and enter the brain, we conducted ex vivo staining studies.
Thus, 5a and 5c with a dose of 40 mg/kg were administered via
tail vein injection to wild-type (WT) mice; 30−60 min after the
injection, the mice were perfused with saline followed by 4%
paraformaldehyde (PFA) and their brains and spinal cords were
then dissected and sectioned. Figure 5 illustrates that the two
lead compounds (5a and 5c) readily entered the brain and
specifically bound to myelin-rich areas of the white matter
regions such as the corpus callosum and striatum. Direct
comparison of fluorescent intensity, albeit qualitative, showed
that such in situ staining with compound 5a (Figure 5A,B) is
higher than with compound 5c (Figure 5C,D).
The lipophilicity in terms of log P of the newly designed
compounds was estimated to be in the range of 4.1−5.3 using
ChemDraw professional 15.0, as listed in Table 1. This range of
lipophilicity is considered high enough to be CNS-permeable,
which thus warrants subsequent synthesis and investigation to
image myelin in vivo.
The synthesis of these novel compounds is shown in Scheme
1. Starting with diethyl (4-aminobenzyl)phosphonate (1), Boc
was used to protect the amino group to generate tert-butyl (4-
((diethoxyphosphoryl)methyl)phenyl)-carbamate (2) in 95%
yield, which was followed by alkylation with 1-bromo-5-
fluoropentane to obtain tert-butyl (4-((diethoxyphosphoryl)-
methyl)phenyl)(5-fluoropentyl)-carbamate (3) in 43% yield.
Compound 3 was then coupled with 4-substituted benzaldehyde
(a−f) through the Horner−Wadsworth−Emmons reaction to
produce intermediates 4a−f in 50−90% yield. A one-pot
reaction was then done using SnCl2 in methanol and ethyl
acetate for 4a and HCl (1.2 M) for 4b−f, which allowed for
reduction of the nitro group to the amino group and
simultaneous deprotection of the Boc group in 50−90% yield
to generate the final products 5a−f, which were used as
radiochemical synthesis standards and for biological studies.23,27
For the radiosynthesis of [18F]5a, the ditosylated compound 7
was prepared from 1,5-diol pentane (6) in 65% yield. As shown
in Scheme 2, the Boc-protected amine of compound 2 was
alkylated in DMF and NaH (60%) to obtain tosylated
compound 8 in 32% yield. The Boc-protected-N-pentane-
tosylated diethyl benzylphosphonate (8) was coupled with 4-
In Vitro Tissue Staining of Focal Demyelination. Based
on the selective staining of 5a in situ in white matter regions, we
further evaluated the capability of compound 5a to detect
myelin changes in a rat model in which demyelination was
induced by injections of lysolethicin unilaterally into the corpus
callosum to generate a focal demyelinated lesion. As shown in
Figure 5E, the presence of demyelinated foci was readily
detected following fluorescent staining of the brain tissue
sections with compound 5a, which was subsequently verified
C
DOI: 10.1021/acs.jmedchem.8b01354
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Scheme 1. Synthesis of [19F]-C5H10-DAS Analogues
a
Reagents and conditions: (i) (Boc)2O, tetrahydrofuran (THF)/H2O (3:1); 20 °C; 24 h; yield, 95%. (ii) NaH(60%); dimethylformamide (DMF);
0 °C; 12 h; yield, 43%. (iii) NaH(60%); DMF; 0 °C; 4 h; yield, 40−80%. (iv) SnCl2/HCl (1.2 M); 70 °C; 6 h; yield, 60−90%.
a
Scheme 2. Radiosynthesis of [18F]5a
a
Reagents and conditions: (i) TsCl, dichloromethane (DCM); 20−0 °C; 12 h; yield, 65%. (ii) THF; NaH (60%); DMF; 0 °C; 6 h; yield, 32%.
(iii) THF (60%); DMF; 0 °C; 4 h; yield, 40%. (iv) [18F]fluoride ion; K222; K2CO3; MeCN; yield, 60%. (v) SnCl2; EtOH; HCl; NaOH; yield, 40%.
using conventional luxol fast blue (LFB) and cresyl violet
staining in adjacent sections (Figure 5F).16
Radiosynthesis. The preceding results from the in vitro and
ex vivo studies encouraged us to evaluate compound 5a for the
in vivo pharmacokinetics profile after labeling with the positron
emitter fluorine-18. A nucleophilic substitution reaction with
fluorine-18 generated by an onsite cyclotron was used to
conduct the radiosynthesis. As shown in Scheme 2, a tosylated
precursor (9a) was first synthesized and subsequently used for
nucleophilic substitution reaction with [18F]KF in the presence
of K2CO3 and kryptofix (K222) in MeCN at 115 °C for 10 min.
Purification of the fluorinated intermediate to remove unreacted
D
DOI: 10.1021/acs.jmedchem.8b01354
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 2. UV absorption spectra of compounds 5a−f.
Figure 5. Ex vivo staining of whole brain and spinal cord tissue sections
with compounds (A, B) 5a and (C, D) 5c. In vitro compound 5a (E)
staining of whole brain tissue section from lysophosphatidylcholine
(LPC)-treated rat that is consistent with luxol fast blue (LFB) staining
in adjacent tissue section (F); demyelinated lesions are shown by the
arrow.
the final compound was determined via HPLC by co-injection of
the compound with a nonradioactive standard. Analytical radio-
HPLC was used to determine the radiochemical purity (RCP) of
the final product, which was >98%. The molar activity (Am) at
the end of synthesis was 92.5 GBq/μmol.
Figure 3. Fluorescence emission and excitation spectra for compounds
5a−f in acetonitrile (1 mM) solution.
In Vitro Autoradiography. Following radiolabeling with F-
18, we conducted autoradiography of mouse brain tissue
sections using compound [18F]5a. Thus, fresh frozen mouse
brain sections were incubated with [18F]5a (2 μCi, 5% ethanol
in phosphate-buffered saline (PBS)). From Figure 6A, it is
evident that distinct labeling of the corpus callosum was
observed after mouse brain sections were exposed to [18F]5a for
20 min, which indicates that the autoradiographic results
correlate with the histological staining results of the myelinated
region. To further illustrate the binding specificity of [18F]5a
Figure 4. FIR in white matter vs gray matter. (A) Representative in
vitro tissue staining showing ROI used for calculation of FIR for white
matter and gray matter. (B) ImageJ used to calculate FIR of each target
compound (5a−f). For each compound, n = 15 sections were used.
free fluorine-18 was done by passing through a silica Sep-Pak.
The 4-nitro group of the F-18-labeled intermediate was then
reduced by SnCl2 in ethanol (EtOH) at 120 °C for 10 min,
followed by acid hydrolysis to cleave the Boc group to yield the
primary amine ([18F]5a). The reaction mixture was cooled to
room temperature, and the pH was adjusted to 8−9 by addition
of sodium hydroxide (NaOH, 1.0 M, 0.5 mL), which was
followed by purification through semipreparative high-perform-
ance liquid chromatography (HPLC) to yield the pure product
[18F]5a with a modest yield of 40%.32 The identity and purity of
Figure 6. In vitro film autoradiography. (A) Specific binding of [18F]5a
to the myelinated corpus callosum in the mouse brain (coronal). The
arrows show myelinated corpus callosum labeled by [18F]5a. (B) No
distinct autoradiographic visualization of the corpus callosum was
observed after pretreatment of the mouse sections (n = 5) with
unlabeled MeDAS (1 mM), indicating that [18F]5a binds to myelinated
corpus callosum with high selectivity and specificity.
E
DOI: 10.1021/acs.jmedchem.8b01354
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 7. (A) Three-dimensional box representation of (A) axial, (S) sagital, and (C) coronal microPET/CT images of a rat brain and cervical spinal
cord following intravenous (i.v.) administration of [18F]5a, showing high uptake in the myelin-rich white matter regions of the brain and spinal cord (n
= 3). (B) Comparison of [18F]5a uptake in WT rat brain and spinal cord with previously developed [18F]TAFDAS and [11C]MeDAS, showing
relatively higher uptake in the brain compared to [11C]MeDAS and significantly higher uptake in spinal cord than with [18F]TAFDAS and
[11C]MeDAS at later time points.
toward myelin, fresh frozen mouse brain sections were
incubated with excess MeDAS, a compound that binds to
myelin with high specificity. After 10 min, [18F]5a (2μCi, 5%
ethanol in PBS) was added to the above solution and the brain
sections were allowed to incubate for another 20 min. As shown
in Figure 6B, pretreatment of the sections with an excess amount
of MeDAS significantly reduced the autoradiographic labeling of
the corpus callosum. After 20 min exposure to [18F]5a, no
distinct labeling of corpus callosum was observed, indicating that
[18F]5a binds to myelinated corpus callosum with high
specificity.
Quantitative MicroPET/CT Imaging of Myelin in the
Brain. Post-radiolabeling, we obtained the pharmacokinetic
profile of [18F]5a in the brain of Sprague-Dawley (SD) rats (n =
3). Quantitative analysis required registering microPET images
to the CT images, as shown in Figure 7A in a three-dimensional
(3D) presentation of axial, coronal, and sagital views to define
ROIs accurately and to quantify the radioactivity concentration
of [18F]5a. The in vivo radioactivity concentration was evaluated
in terms of average standardized uptake value (SUV) in the
spinal cord and brain. [18F]5a showed high initial uptake in the
brain, which peaked at 1 min post-injection and was followed by
rapid clearance before reaching a stable plateau at 20 min post-
injection. This profile suggests that [18F]5a is a promising
myelin-imaging radioligand with high brain permeability and
binding affinity for myelin. As shown in Figure 7B, compared to
previously developed [18F]TAFDAS and [11C]MeDAS, [18F]5a
exhibited the highest uptake in the spinal cord, and the uptake of
[18F]5a in the brain was similar to C-11-labeled [11C]MeDAS
but significantly higher than [18F]TAFDAS at later time points.
The in vivo binding specificity of [18F]5a was further
evaluated by investigating the pharmacokinetic profiles of
[18F]5a in a shiverer mouse model that is myelin-deficient in
the brain. For direct comparison, both shiverer mice and their
age-matched littermates were imaged side by side in a PET/CT
scanner. After [18F]5a was administered, dynamic emission
scans were immediately acquired for 60 min in 3D list mode.
The brain uptake of [18F]5a was quantified after co-registration
of PET images with CT images. The radioactivity in terms of
percent of injected dose (%ID) between 35 and 60 min was
summed and compared after the brain concentration of [18F]5a
reached a plateau. As shown in Figure 8, the average uptake
(35−60 min) of [18F]5a in the shiverer mouse brain was 10.87
0.72% ID/g, which was significantly decreased compared to the
Figure 8. Average brain uptake of [18F]5a between 35 and 60 min post-
injection in shiverer mice (blue) and age-matched control (red) (p =
0.0416, two-tailed t test).
Ex Vivo Three-Dimensional Cryofluorescence Imag-
ing. To further validate the in vivo PET imaging results, we
conducted ex vivo 3D cryofluorescence imaging of the rat spinal
cord in the same animals following [18F]5a-PET/CT scans.
Thus, sequential slices (100 μm thickness) of rat spinal cord
were first obtained with a cryo-microtome 30 min after tail vain
injection of compound 5a after the [18F]5a-PET/CT scan.
Fluorescent images were then captured at 10 μm pixel resolution
with a fluorescence microscope (Leica DM4000B, Leica
Microsystems Inc, Buffalo Grove, IL) to generate an image
volume composed of the stacked sequential slices. Misalignment
in Cryo-3D images was reduced by aligning sequential slices
with control point registration and a rigid-body transformation
in Matlab. The Cryo-3D image stack was then visualized using
the Amira software package and aligned with microPET/CT
images (Figure 9). Cryo-3D provides high-resolution ex vivo
data, which could be analyzed from multiple views and color
map representations (Figure 10).
Quantitative MicroPET/CT Imaging Studies in the
Spinal Cord. The aforementioned studies of [18F]5a PET/CT
imaging of myelin in the brain were followed by conducting
[18F]5a PET/CT imaging of the spinal cord in a rat model of
SCI, which involved making a contusion of the spine at thoracic
region T12 to introduce demyelination. A baseline scan prior to
and 1 day after the contusion was done with [18F]5a PET
control littermates (19.39
1.50% ID/g, p= 0.0416, two-
tailed t test).
F
DOI: 10.1021/acs.jmedchem.8b01354
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 9. MicroPET/CT/Cryo-3D fluorescence co-registered images: (A) axial, (B) coronal, (C) sagittal, and (D) 3D view of the rat brain and spinal
cord following intravenous administration of [18F]5a showing high uptake in the myelin-rich white matter region of the brain. Cryo-3D fluorescence is
shown in the T1−13 region of the spinal cord.
Figure 10. Ex vivo cryogenic 3D images of axial and sagital views and color map processed with Amira for the spinal cord sections (n = 5) using
compound 5a.
imaging. [18F]5a uptake in four spine segments (T10−13) was
determined and normalized to the average (T10−13) spine
uptake. As shown in Figure 11A,B, the intact thoracic region of
the spinal cord before contusion was clearly visualized by
[18F]5a PET imaging. After contusion, the lesion at vertebrae
level T12 was readily detected by [18F]5a PET (compare Figure
11E,F). Figure 11E illustrates that the uptake of [18F]5a in T12
in the SCI group was 0.86, which was significantly lower than
that in the baseline scans (1.00). Post-microPET/CT imaging,
demyelination in the lesion region was verified by in situ
histological staining. As shown in Figure 11F−H, 30 min after
injection of nonlabeled compound 5a through the tail vein, no
staining was observed in the dorsal column after contusion,
which was further verified by LFB and cresyl violet staining.
DISCUSSION AND CONCLUSIONS
■
Discovery and development of drugs for biomedical imaging
represent a new frontier in medicinal chemistry. Compared to
therapeutic drugs, imaging drugs are used in much smaller
quantities, at least 1000-fold less than therapeutic drugs.
Consequently, imaging drugs require less rigorous toxicity
testing but more stringent requirements for target-binding
specificity, in vivo pharmacokinetic properties, and metabolic
activities. Probably the most distinctively important aspect in
imaging drug development lies in the structural design of each
particular imaging modality even for imaging the same target.
Each different imaging modality requires a specific signaling
element and thus distinctive structural design.
Figure 11. Representative PET/CT fusion images in rats acquired
before SCI surgery (baseline scan) with higher-magnification coronal
(A) and sagittal (B) images in the T12 spinal cord. Representative
PET/CT fusion images in rats acquired 1 day after SCI surgery with
higher-magnification coronal (E) and sagittal (F) images in the T12
spinal cord. Note the decreased signal contrast in the SCI animal (n = 3)
compared to the baseline scan. In situ histological staining of the SCI
(T12) tissue section with reference compound 5a after microPET/CT
imaging showing a demyelinated lesion at the dorsal portion (H), which
is consistent with LFB staining using adjacent sections (G). (I)
Quantification of the total cumulative [18F]5a uptake in the T10−13 in
terms of normalized SUV ratio at 30−60 min post-injection showing
significantly lower uptake in the SCI (T12) compared to that in baseline
scans.
In this work, we took a distinct approach to structural design
and development of a new class of myelin-imaging agents that
are compatible with PET imaging and 3D cryoimaging. The two
imaging modalities complement each other, i.e., PET can be
used to detect and quantify myelin distribution in vivo, whereas
3D cryoimaging can be used to characterize microscopically the
myelin histology in vitro. Since both imaging modalities are
tomographic, PET images and 3D cryoimages can be co-
registered for cross-referencing and validation. This requires use
G
DOI: 10.1021/acs.jmedchem.8b01354
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
of a structurally identical myelin-imaging agent for both imaging
modalities.
deficient shiverer mice decreased significantly. Thus, high
binding specificity for myelin was observed in vitro, ex vivo,
and in vivo in a series studies showing either labeled or
nonlabeled 5a bound selectively to myelin in the presence of
other proteins at the tissue level. In addition, following tail vein
injection of [18F]5a, no sign of acute toxicity in rats was
observed. These studies suggest that [18F]5a can be used to
detect myelin changes in the brain.
For this reason, we designed and synthesized a series of
stilbene derivatives that are capable of radiolabeling with F-18
for PET and fluorescent staining for 3D cryoimaging. For SAR
studies, we first focused on lipophilicity, which is an important
physicochemical property for both histochemical staining and in
vivo pharmacokinetics in the CNS. A total of six compounds
were then designed by introducing NO2, CH3, SCH3, OCH3,
NHCH3, and NH2 to one phenyl ring and fluoropentyl amino to
the other phenyl ring of stilbene. Following synthesis, the
lipophilicity varies as different functional groups were
introduced to one of the phenyl rings, which decreases in the
order NO2, CH3, SCH3, OCH3, NHCH3, and NH2. For
permeability across the blood−CNS barrier, myelin-imaging
agents ought to exhibit moderate lipophilicity. But high
lipophilicity often results in high nonspecific binding.
In addition, the fluorescent properties of these compounds
were also characterized, which is shown in Figure 3. The
compounds are all seen to be fluorescent with absorption and
excitation wavelengths of ∼400 nm. Among all of the
compounds tested, compounds 5a and 5e showed the highest
fluorescent intensity. Subsequent in vitro fluorescent tissue
staining was then conducted to further evaluate the test
compounds. For quantitative SAR studies, an FIR was defined
between the fluorescent intensity in a myelin-rich genu corpus
callosum region and a myelin-deficient subcortical gray matter
region. The higher FIR suggests high binding specificity and vice
versa. As shown in Figure 4, compounds 5a and 5c showed the
highest FIR values among all of the test compounds. These in
vitro studies led us to identify compounds 5a, 5c, and 5e as lead
compounds. Considering that the lipophilicity of compound 5e
is abnormally high, compounds 5a and 5c were thus selected for
downstream ex vitro and in vivo studies.
Thanks to the fluorescence exhibited by compounds 5a and
5c, we were able to first evaluate the spinal cord and brain uptake
and binding specificity for myelin. After tail vein injection, both
compounds readily entered the brain and selectively bound to
myelin in the brain and spinal cord. Interestingly, even though
both compounds have similar cLog P values, compound 5a
displayed potent staining with higher fluorescent intensity than
compound 5c, as shown in Figure 5A−D. These in vitro and ex
vivo studies suggest that all of the test compounds bind
selectively to myelin in the presence of other proteins at the
tissue level. This class of compounds binds to myelin not
through lipid−lipid interaction but rather through molecular
interaction, although the exact binding mechanism has not been
defined and needs to be further investigated. As a result,
compound 5a could readily detect myelin-damaged lesions
induced by lysophosphatidylcholine (LPC) in a rat model, as
shown in Figure 5E,F. In addition, following tail vein injection of
either nonlabeled compound 5a at a dose as high as 40 mg/kg,
no sign of acute toxicity in rats was observed.
The aforementioned studies prompted us to further
investigate compound 5a as a radiotracer for PET imaging of
myelin. Following radiolabeling with F-18, we tested [18F]5a in
two animal models. PET imaging was first conducted in rats
showing [18F]5a readily entered the spinal cord and brain. The
uptake was significantly higher than that of previously developed
myelin-imaging agents such as MeDAS and TAFDAS. To
determine the binding specificity for myelin in vivo, [18F]5a was
further evaluated in the shiverer mouse model. Compared to
age-matched control littermates, brain uptake in myelin-
Following PET studies, histological analysis is required to
validate the imaging results. Conventional immunohistochem-
istry is often conducted in selected tissue sections, which is
limited to analysis of small fragments of the brain and spinal cord
and is inherently biased due to the limited scope of sampling.
Three-dimensional cryoimaging, on the other hand, provides a
tomographical characterization of the whole tissue of interest
and thus is similar to PET imaging. The 3D cryoimaging results
are therefore less subjective than conventional immunohisto-
chemistry as sampling bias is essentially eliminated. In addition,
thanks to the inherent fluorescence of compound 5a and its
optimal pharmacokinetics, histological 3D cryoimaging can also
be conducted in situ, which provides the same tomographical
characterization of myelin distribution at a microscopic
resolution that can be co-registered with PET images for precise
validation of the PET imaging results, as shown in Figures 9 and
10. Finally, the imaging sensitivity and specificity of [18F]5a-
PET was ultimately determined in an SCI model with focal
demyelination (Figure 11). After contusion at vertebrae level
T12, the uptake of [18F]5a was significantly decreased compared
to the baseline scan due to myelin damage, which was further
confirmed by in situ staining and histochemistry by LFB and
cresyl violet staining.
In summary, a multimodal approach combining in vivo PET
and ex vivo 3D cryoimaging has been developed for optimizing
the selection of novel F-18-labeled radioligands for imaging
myelin. Following the design and synthesis of a series of amino-
pentyl-fluorinated stilbene derivatives, SAR studies were
conducted for characterization and optimization of in vitro
and ex vivo binding properties, which led to identification of a
lead compound 5a that selectively binds to myelin and readily
enters the brain and spinal cord. Subsequent in vivo PET
imaging studies in a rat model of contusive SCI showed that
[18F]5a-PET can readily detect myelin damage in the spinal
cord, which suggests that [18F]5a is a potentially valuable PET
imaging agent for monitoring myelin changes in the CNS.
EXPERIMENTAL SECTION
■
General Procedures. All chemicals and reagents were used as
received without further purification. Glassware was dried in an oven at
130 °C and purged with dry air before use. Unless otherwise mentioned,
reactions were performed open to air. The reactions were monitored by
thin-layer chromatography (TLC) and monitored by a dual short-/
long-wavelength UV lamp. Flash column chromatography was
performed using 230−400 mesh silica gel (Fisher). Nuclear magnetic
resonance (NMR) spectra were recorded on a Varian INOVA 400
spectrometer and a 500 MHz Bruker Ascend Avance III HD at room
1
temperature. Chemical shifts for H, 13C, and 19F NMR spectra were
reported as δ, parts per million, and referenced to an internal deuterated
solvent central line. Multiplicity and coupling constants (J) were
calculated automatically on MestReNova 10.0, NMR processing
software from Mestrelab Research. The purity of the newly synthesized
compounds as determined by analytical high-performance liquid
chromatography (HPLC) was >95% on C-18 reversed-phase HPLC
(Phenomenex, 10 mm × 250 mm) using an eluent of MeCN/H2O =
60:40 at a flow rate of 3.0 mL/min. High-resolution mass spectrometry-
electrospray ionization (HRMS-ESI) images were acquired on an
H
DOI: 10.1021/acs.jmedchem.8b01354
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Agilent Q-TOF. UV absorption was measured on a Cary 50 Bio
spectrophotometer using a standard 1 cm × 1 cm quartz cuvette.
Fluorescence was measured with a Cary Eclipse spectrophotometer
using a 1 cm × 1 cm quartz cuvette in a 10 mM MeCN solution.
tert-Butyl (4-((Diethoxyphosphoryl)methyl)phenyl)-
carbamate (2). Compound 2 was synthesized according to our earlier
published method.23 To a 100 mL round-bottom flask with a magnetic
stir bar, diethyl (4-aminobenzyl)phosphonate (1, 5.0 g, 20.56 mmol),
di-tert-butyl dicarbonate (4.50 g, 21.0 mmol), tetrahydrofuran (THF,
25 mL), and water (10 mL) were added. The reaction mixture was
stirred at room temperature open to air overnight. After completion of
the reaction, THF was evaporated under vacuum and the resulting
residue was diluted with water and extracted with ethyl acetate (EtOAc,
50 mL × 3). The organic layers were combined and washed with water
(50 mL × 2) followed by brine (50 mL). The organic layer was dried
over magnesium sulfate (MgSO4), filtered, and evaporated under
reduced pressure. A white amorphous compound was obtained (6.5 g;
yield, 95%) and used without further purification.23 1H NMR (400
MHz, chloroform-d) δ 7.30 (br d, J = 8.4 Hz, 2H), 7.17−7.19 (m, 2H),
6.87 (br s, 1H), 4.04−3.91 (m, 4H), 3.07 (d, 2JH,P = 21.2 Hz, 2H), 1.49
(s, 9H), 1.22 (td, 3JH,H = 6.8 Hz, 4JH,P = 0.4 Hz, 6H). 13C NMR (100
MHz, chloroform-d) δ 152.8, 137.4, 130.1, 125.5, 118.4, 80.3, 62.0, 32.9
(d, 1JC,P = 138 Hz, 1C), 28.8, 16.3.
tert-Butyl (E)-(5-Fluoropentyl)(4-(4-nitrostyryl)phenyl)-
carbamate (4a and 4b). The yellow sticky oil was purified in 10%
EtOAc/hexane, yielding 0.25 g (79%). 1H NMR (400 MHz,
chloroform-d) δ 8.22 (d, J = 8.5 Hz, 2H), 7.63 (d, J = 8.8 Hz, 2H),
7.52 (d, J = 8.6 Hz, 2H), 7.28−7.21 (m, 3H), 7.11 (d, J = 15.2 Hz, 1H),
4.54−4.45 (m, 1H), 4.41−4.33 (m, 1H), 3.67 (t, J = 7.4 Hz, 2H), 1.77−
1.52 (m, 6H), 1.45 (s, 9H). 13C NMR (101 MHz, chloroform-d) δ
168.3, 154.7, 144.0, 143.1, 133.9, 132.7, 127.5, 127.3, 127.0, 126.4,
124.4, 84.9, 83.3, 80.6, 49.8, 30.3, 30.1, 28.5, 28.3, 22.6, 22.6. 19F NMR
(376 MHz, chloroform-d) δ −85.16 to −86.41 (m), −100.01.
tert-Butyl (E)-(5-Fluoropentyl)(4-(4-(methylamino)styryl)-
phenyl)carbamate (4c). The light yellow precipitate (ppt.) was
obtained after water quenching and washing with water three times (20
mL each) and hexane once (20 mL), then dried under vacuum, yielding
0.080 g (46%). 1H NMR (400 MHz, chloroform-d) δ 7.42 (d, J = 8.1
Hz, 2H), 7.36 (d, J = 8.2 Hz, 2H), 7.12 (d, J = 8.0 Hz, 2H), 7.00 (d, J =
16.2 Hz, 1H), 6.88 (d, J = 16.1 Hz, 1H), 6.60 (d, J = 8.2 Hz, 2H), 4.47 (t,
J = 6.1 Hz, 1H), 4.35 (t, J = 6.1 Hz, 1H), 3.64 (t, J = 7.5 Hz, 2H), 2.87 (s,
3H), 1.68 (dd, J = 24.9, 7.4 Hz, 4H), 1.57−1.51 (m, 2H), 1.43 (s, 9H).
13C NMR (101 MHz, chloroform-d) δ 149.2, 141.1, 136.1, 136.1, 129.1,
127.9, 127.3, 126.4, 123.8, 112.6, 85.0, 80.3, 49.8, 30.8, 30.3, 30.1, 28.5,
28.2, 22.6. 19F NMR (376 MHz, chloroform-d) δ −100.01.
tert-Butyl (E)-(5-Fluoropentyl)(4-(4-(methylthio)styryl)-
phenyl)carbamate (4d). The white ppt. was obtained after water
quenching, washed with water three times (20 mL each), and dried
under vacuum, yielding 0.20 g (63%). 1H NMR (400 MHz, chloroform-
d) δ 7.46 (d, J = 8.5 Hz, 2H), 7.43 (d, J = 8.4 Hz, 2H), 7.24 (d, J = 8.4
Hz, 2H), 7.16 (d, J = 8.1 Hz, 2H), 7.03 (d, J = 1.5 Hz, 2H), 4.47 (t, J =
6.1 Hz, 1H), 4.36 (t, J = 6.1 Hz, 1H), 3.68−3.62 (m, 2H), 2.51 (s, 3H),
1.76−1.55 (m, 6H), 1.44 (s, 9H). 13C NMR (101 MHz, chloroform-d)
δ 154.5, 141.5, 137.7, 134.8, 134.0, 127.8, 127.1, 126.9, 126.7, 126.5,
94.2, 84.6, 82.9, 80.0, 49.4, 30.0, 29.8, 28.2, 27.9, 22.2, 22.2, 15.6. 19F
NMR (376 MHz, chloroform-d) δ −84.26 to −88.06 (m), −100.01.
tert-Butyl (E)-(5-Fluoropentyl)(4-(4-methoxystyryl)phenyl)-
carbamate (4e). The white amorphous compound was obtained
from a flash silica column in 10% EtOAc/hexane in a yield of 0.10 g
(36%). 1H NMR (400 MHz, chloroform-d) δ 7.46 (d, J = 1.9 Hz, 2H),
7.44 (d, J = 1.5 Hz, 2H), 7.15 (d, J = 8.1 Hz, 2H), 7.01 (s, 1H), 6.97 (s,
1H), 6.92−6.88 (m, 2H), 4.47 (t, J = 6.1 Hz, 1H), 4.36 (t, J = 6.1 Hz,
1H), 3.83 (s, 3H), 3.69−3.61 (m, 2H), 1.77−1.58 (m, 4H), 1.57 (d, J =
1.8 Hz, 2H), 1.43 (d, J = 2.9 Hz, 9H). 13C NMR (101 MHz,
chloroform-d) δ 159.4, 154.8, 141.5, 135.5, 130.2, 128.4, 127.8, 127.5,
126.6, 125.0, 114.3, 84.9, 83.3, 80.3, 55.5, 49.8, 30.3, 30.1, 28.5, 28.2,
28.1, 22.5, 22.5. 19F NMR (376 MHz, chloroform-d) δ −85.01 to
−86.81 (m), −100.01.
tert-Butyl (4-((Diethoxyphosphoryl)methyl)phenyl)(5-
fluoropentyl)carbamate (3). To a 100 mL round-bottom flask
purged with argon gas and fitted with a magnetic stirrer were added
sodium hydride (NaH, 0.174 g, 4.37 mmol, 60%) and tert-butyl (4-
((diethoxyphosphoryl)methyl)phenyl)carbamate (2, 0.50 g, 1.46
mmol). The mixture was purged with argon gas, and dry
dimethylformamide (DMF, 4 mL) at 0 °C was added. To the reaction
mixture, 1-bromo-5-fluoropentane (0.4 g, 2.91 mmol, d = 1.31 g/mL)
was added dropwise after 30 min at 0 °C under argon gas. The reaction
was stirred under argon and allowed to reach room temperature
overnight. After completion, the reaction was quenched with water. The
residue was dissolved in EtOAc and the aqueous layer was extracted
with EtOAc (30 mL × 3). The organic layers were combined and
washed with water (50 mL × 2) and brine (50 mL). The organic layer
was dried over MgSO4 and then filtered and concentrated to yield the
desired product (3) as a sticky oil. This was purified over a silica flash
column using 30% EtOAc in hexane as eluent, which resulted in the
pure compound (0.55 g; yield, 43%). 1H NMR (400 MHz, chloroform-
d) δ 7.29 (d, J = 2.6 Hz, 1H), 7.27−7.26 (m, 1H), 7.11 (d, J = 8.0 Hz,
2H), 4.47 (t, J = 6.1 Hz, 1H), 4.36 (t, J = 6.1 Hz, 1H), 4.07−3.95 (m,
4H), 3.67−3.58 (m, 2H), 3.15 (d, 2JH,P = 21.2 Hz, 2H), 1.76−1.61 (m,
4H), 1.57 (q, J = 7.3 Hz, 2H), 1.41 (s, 9H), 1.26−1.22 (m, 6H). 13C
NMR (101 MHz, chloroform-d) δ 141.4, 130.4, 130.3, 129.5, 127.4,
tert-Butyl (E)-(5-Fluoropentyl)(4-(4-methylstyryl)phenyl)-
carbamate (4f). The white precipitate was obtained after water
quenching and washing with water three times under vacuum (20 mL
each); this compound was dried under high vacuum, yielding 0.25 g
(83%). 1H NMR (500 MHz, chloroform-d) δ 7.46 (d, J = 8.1 Hz, 2H),
7.41 (d, J = 7.8 Hz, 2H), 7.16 (t, J = 7.1 Hz, 4H), 7.04 (s, 2H), 4.46 (t, J
= 6.1 Hz, 1H), 4.37 (t, J = 6.1 Hz, 1H), 3.65 (t, J = 7.4 Hz, 2H), 2.36 (s,
3H), 1.69 (dt, J = 25.2, 7.5 Hz, 2H), 1.61−1.51 (m, 4H), 1.43 (s, 9H).
13C NMR (126 MHz, chloroform-d) δ 154.8, 141.7, 137.7, 135.3, 134.6,
1
84.9, 83.3, 80.3, 62.4, 62.32, 49.8, 34.2, 32.8 (d, JC,P = 138 Hz, 1C),
30.3, 30.1, 28.5, 28.5, 28.2, 22.6, 22.5, 16.6, 16.5. 19F NMR (376 MHz,
chloroform-d) δ −86.02, −100.01.
General Method for the Synthesis of 4a−f (Scheme 1). To a
100 mL round-bottom flask purged with argon and fitted with a
magnetic stirrer was added sodium hydride (2.0 equiv, 60% dispersion
in mineral oil). The flask was purged with argon, and 2.0 mL of dry
DMF was added. The compound (3, 1.1 equiv) was dissolved in dry
DMF (2.0 mL) and transferred via syringe in a NaH−DMF mixture to a
flask. The mixture was stirred under argon at 0 °C for 1 h. Commercially
procured para-benzaldehyde (a−f, 1.0 equiv) was dissolved in dry
DMF (2.0 mL) and added to the reaction mixture via a syringe under
argon at 0 °C. The reaction mixture was stirred for 3 h at 0 °C in the
dark under argon. The reactions were quenched with water and some of
the derivatives precipitated out as an amorphous solid, which was
filtered under vacuum and washed three to four times with water (50
mL each). Some of the compounds were extracted with EtOAc (30 mL
× 3). The organic layers were combined and washed with water (25 mL
× 2) followed by brine (25 mL). The organic layer was dried over
MgSO4, filtered, and evaporated under reduced pressure to yield the
crude products. These were purified by dissolving in a minimal amount
of dichloromethane (DCM) and made into a slurry with silica gel and
then dried under vacuum. The compound was eluted with 10−15%
EtOAc in hexane in a flash column.
129.5, 128.8, 127.3, 127.1, 126.8, 126.5, 84.7, 83.4, 80.3, 49.8, 30.3,
30.1, 28.5, 28.2, 28.1, 22.5, 22.5, 21.4. 19F NMR (376 MHz, chloroform-
d) δ −85.01 to −86.81 (m), −100.01.
General Method for Boc-Deprotection and Reduction (5a−
f). In a 100 mL round-bottom flask fitted with a stir bar, compounds
4a−f were dissolved in EtOAc (15 mL) and methanol (10 mL).
Stannous chloride (SnCl2, 10 equiv)/hydrochloric acid (HCl, 1.2 M, 2
mL) was added to the solution. The mixture was fitted with a water
condenser and heated to 70 °C for 6 h open to air in the dark. The
reactions were monitored through thin-layer chromatography (TLC);
after completion, the solvent was removed under vacuum. The residue
was quenched by aqueous sodium bicarbonate (NaHCO3, 20%) until
bubbles stopped forming. Some of the compounds were precipitated
and filtered under vacuum and washed with water three to four times
(50 mL each). The other compounds were extracted in EtOAc (30 mL
× 3) and washed with aqueous NaHCO3 (20%, 30 mL × 2) and once
I
DOI: 10.1021/acs.jmedchem.8b01354
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
with water (50 mL) followed by brine (50 mL). It was then dried over
MgSO4 and the solvent was evaporated under reduced pressure. The
crude product was dissolved in a minimal DCM slurry on silica and
purified on silica by flash column chromatography using 10−15%
EtOAc in hexane.22
(E)-4-(4-Aminostyryl)-N-(5-fluoropentyl)aniline (5a). The yel-
low amorphous compound was eluted from a silica flash column using
10% EtOAc/hexane, yield 0.080 g (45%). 1H NMR (400 MHz,
chloroform-d) δ 7.30 (t, J = 8.5 Hz, 4H), 6.83 (d, J = 2.2 Hz, 2H), 6.66
(d, J = 8.5 Hz, 2H), 6.58 (d, J = 8.6 Hz, 2H), 4.53 (t, J = 6.0 Hz, 1H),
4.41 (t, J = 6.0 Hz, 1H), 3.71 (s, 2H), 3.16 (t, J = 7.0 Hz, 2H), 1.77 (d, J
= 8.0 Hz, 1H), 1.77−1.63 (m, 4H), 1.61−1.49 (m, 3H). 13C NMR (101
MHz, chloroform-d) δ 147.7, 145.5, 129.1, 127.6, 127.5, 127.4, 125.6,
124.7, 115.5, 113.0, 85.0, 83.4, 44.0, 30.5, 30.3, 29.4, 23.1, 23.1. 19F
NMR (376 MHz, chloroform-d) δ −85.52 to −86.41 (m), −100.01.
HRMS (ESI+) calcd for (C19H23FN2) [M + H]+ 299.1918, found
299.1919. HPLC purity: 99.99%, retention time: 9.53 min.
(E)-N-(5-Fluoropentyl)-4-(4-nitrostyryl)aniline (5b). The or-
ange precipitate was obtained after quenching, in a yield of 0.080 g
(65%). 1H NMR (400 MHz, chloroform-d) δ 8.18 (d, J = 8.8 Hz, 2H),
7.55 (d, J = 8.8 Hz, 2H), 7.39 (d, J = 8.5 Hz, 2H), 7.19 (d, J = 16.2 Hz,
1H), 6.91 (d, J = 16.2 Hz, 1H), 6.60 (d, J = 8.6 Hz, 2H), 4.54 (t, J = 5.9
Hz, 1H), 4.42 (t, J = 5.9 Hz, 1H), 3.90 (s, 1H), 3.19 (t, J = 7.0 Hz, 2H),
1.83−1.65 (m, 4H), 1.59−1.51 (m, 3H). 13C NMR (101 MHz,
chloroform-d) δ 148.8, 144.8, 133.5, 128.4, 125.9, 125.1, 124.0, 121.4,
112.4, 84.6, 82.9, 43.3, 30.1, 29.9, 28.9, 22.8, 22.7. 19F NMR (376 MHz,
chloroform-d) δ −85.01 to −86.96 (m), −100.01. HRMS (ESI+) calcd
for (C19H21FN2O2) [M + H]+ 329.1660, found 329.1662. HPLC
purity: 95.36%, retention time: 7.52 min.
chloroform-d) δ 7.35 (t, J = 8.3 Hz, 4H), 7.13 (d, J = 7.8 Hz, 2H), 6.98
(d, J = 16.3 Hz, 1H), 6.87 (d, J = 16.3 Hz, 1H), 6.60 (d, J = 8.2 Hz, 2H),
4.53 (t, J = 6.0 Hz, 1H), 4.41 (t, J = 6.0 Hz, 1H), 3.16 (t, J = 7.0 Hz, 2H),
2.34 (s, 3H), 1.84−1.63 (m, 5H), 1.54 (q, J = 8.0 Hz, 3H). 13C NMR
(126 MHz, chloroform-d) δ 147.7, 136.7, 135.4, 129.4, 127.9, 127.7,
127.3, 126.1, 124.6, 113.1, 84.1 (d, J = 164.6 Hz), 44.1, 30.4, 30.3, 29.8,
29.2, 23.1, 23.03 21.3. 19F NMR (376 MHz, chloroform-d) δ −84.62 to
−87.35 (m), −100.01. HRMS (ESI+) calcd for (C20H24FN) [M + H]+
298.1966, found 298.1970. HPLC purity: 94.85%, retention time: 8.25
min.
Synthesis of Pentane-1,5-diyl Bis(4-methylbenzenesulfo-
nate) (7). To a 100 mL round-bottom flask with a magnetic stir bar
charged with 4-methylbenzenesulfonyl chloride (2.29 g, 12.0 mmol)
was added 25 mL of dry DCM as solvent, followed by trimethylamine
(4.0 mL, 28.80 mmol). The solution was stirred for 30 min at room
temperature and cooled to 0 °C. After 1 h at 0 °C, 1,5-diol-pentane
(0.50 g, 4.80 mmol) was added and the reaction mixture was stirred
overnight and warmed to room temperature. The reaction was
monitored via TLC, and after completion of the reaction, the solvent
was removed under reduced pressure, quenched with water, and
extracted with EtOAc (50 mL × 3). The combined organic layer was
again washed with water (50 mL × 2) followed by brine (50 mL). The
organic layer was dried over MgSO4 and evaporated under reduced
pressure. The residue was dissolved in DCM/hexane (3:7) and kept at
−10 °C for 4 h to yield a white crystallized compound in a yield of 1.30 g
(65%). 1H NMR (500 MHz, chloroform-d) δ 7.77 (d, J = 8.0 Hz, 4H),
7.35 (d, J = 7.9 Hz, 4H), 3.97 (t, J = 6.3 Hz, 4H), 2.46 (s, 6H), 1.67−
1.56 (m, 4H), 1.42−1.31 (m, 2H). 13C NMR (126 MHz, chloroform-d)
δ 145.0, 133.1, 130.0, 128.0, 70.1, 28.3, 21.8, 21.6.
(E)-N-(5-Fluoropentyl)-4-(4-(methylamino)styryl)aniline
(5c). The white amorphous compound was obtained from a silica flash
column in 15% EtOAc/hexane in a yield of 0.030 g (60%). 1H NMR
(400 MHz, chloroform-d) δ 7.34 (s, 1H), 7.32 (d, J = 2.9 Hz, 2H), 7.30
(s, 1H), 6.83 (s, 2H), 6.60 (s, 1H), 6.58 (d, J = 3.6 Hz, 2H), 6.57 (s,
1H), 4.53 (t, J = 6.0 Hz, 1H), 4.41 (t, J = 6.0 Hz, 1H), 3.16 (t, J = 7.0 Hz,
2H), 2.86 (s, 3H), 1.77 (s, 1H), 1.74−1.64 (m, 3H), 1.59−1.54 (m,
1H), 1.52 (d, J = 6.6 Hz, 1H). 13C NMR (101 MHz, chloroform-d) δ
148.6, 147.5, 127.8, 127.4, 127.3, 125.0, 113.0, 112.7, 84.9, 83.3, 44.0,
31.0, 30.5, 30.3, 29.4, 23.1, 23.1. 19F NMR (376 MHz, chloroform-d) δ
−85.16 to −86.61 (m), −100.01. HRMS (ESI+) calcd for (C20H25FN2)
[M + H]+ 313.2075, found 313.2076. HPLC purity: 95.48%, retention
time: 10.90 min.
(E)-N-(5-Fluoropentyl)-4-(4-(methylthio)styryl)aniline (5d).
The off-brown amorphous precipitate was formed by washing with
water four times (50 mL each), yielding 0.125 g (81%). 1H NMR (400
MHz, chloroform-d) δ 7.42 (d, J = 7.9 Hz, 2H), 7.40−7.32 (m, 4H),
7.18 (d, J = 7.7 Hz, 2H), 6.97 (d, J = 2.3 Hz, 2H), 4.45 (t, J = 6.0 Hz,
1H), 4.33 (t, J = 5.9 Hz, 1H), 3.24 (t, J = 8.0 Hz, 2H), 2.50 (s, 3H), 1.87
(s, 2H), 1.76−1.56 (m, 3H), 1.54−1.42 (m, 2H). 13C NMR (101 MHz,
chloroform-d) δ 138.6, 133.8, 129.3, 127.8, 127.1, 126.7, 126.5, 122.01,
83.7 (d, J = 164.9 Hz), 51.4, 30.1, 29.9, 26.0, 22.7, 22.7, 15.8. 19F NMR
(376 MHz, chloroform-d) δ −86.41 (td, J = 47.6, 23.9 Hz), −100.01.
HRMS (ESI+) calcd for (C20H24FNS) [M + H]+ 330.1686, found
330.1692. HPLC purity: 97.58%, retention time: 7.20 min.
5-((tert-Butoxycarbonyl)(4-((diethoxyphosphoryl)methyl)-
phenyl)amino)pentyl-4-methylbenzenesulfonate (8). To a 100
mL round-bottom flask purged with argon and fitted with a stir bar was
added sodium hydride (0.238 g, 6.00 mmol, 60%) and tert-butyl (4-
((diethoxyphosphoryl)methyl)phenyl)carbamate (2, 1.0 g, 2.91
mmol). The mixture was purged with argon, and dry DMF (5 mL) at
0 °C was added. Pentane-1,5-diyl bis(4-methylbenzenesulfonate) (7,
1.10 g, 2.65 mmol) was added after 30 min at 0 °C under argon. The
reaction was stirred under argon and allowed to reach room
temperature overnight. After completion, the reaction was quenched
with water. The residue was dissolved in EtOAc and water, and the
aqueous layer was extracted with EtOAc (30 mL × 3). The organic
layers were combined and washed with water (50 mL × 2) followed by
brine (50 mL). The organic layer was dried over MgSO4, then filtered
and concentrated under vacuum to yield the desired product 3 as a
sticky oil. This was purified over a silica flash column using 50% EtOAc
in hexane as eluent to get the desired compound as an oil in a yield of
0.50 g (32%). 1H NMR (500 MHz, chloroform-d) δ 7.76 (d, J = 8.3 Hz,
2H), 7.33 (d, J = 8.0 Hz, 2H), 7.25 (d, J = 2.6 Hz, 2H), 7.08 (d, J = 8.0
Hz, 2H), 4.06−3.96 (m, 6H), 3.54 (t, J = 7.4 Hz, 2H), 3.13 (d, JP−H
=
21.6 Hz, 2H), 2.44 (s, 3H), 1.47 (p, J = 9.2, 7.8 Hz, 3H), 1.38 (s, 9H),
1.33−1.26 (m, 3H), 1.23 (t, J = 7.1 Hz, 6H). 13C NMR (126 MHz,
chloroform-d) δ 154.8, 144.8, 141.3, 141.3, 133.3, 130.3, 130.3, 130.0,
129.6, 129.5, 128.0, 127.3, 80.2, 70.5, 62.3, 62.3, 33.4 (d, JC,P = 138.4
Hz, 1C), 28.7, 28.5, 28.4, 27.9, 22.7, 21.8, 16.5, 16.5.
(E)-N-(5-Fluoropentyl)-4-(4-methoxystyryl)aniline (5e). The
off pink ppt. was obtained, washed four times with water (50 mL each)
under vacuum, and dried under high vacuum, yielding 0.070 g (92%).
1H NMR (400 MHz, chloroform-d) δ 7.44 (s, 1H), 7.43−7.38 (m, 2H),
7.35 (d, J = 8.4 Hz, 2H), 6.92−6.84 (m, 4H), 6.77 (d, J = 8.1 Hz, 2H),
4.50 (d, J = 6.1 Hz, 1H), 4.38 (d, J = 6.1 Hz, 1H), 3.82 (s, 3H), 3.18 (t, J
= 7.3 Hz, 2H), 1.71 (dt, J = 13.3, 6.1 Hz, 5H), 1.52 (t, J = 7.1 Hz, 3H).
13C NMR (101 MHz, chloroform-d) δ 130.8, 127.9, 127.6, 127.5, 127.3,
126.7, 126.5, 115.2, 114.3, 114.3, 84.0 (d, J = 164.3 Hz), 55.5, 30.4,
30.2, 28.5, 23.0, 23. 19F NMR (376 MHz, chloroform-d) δ −84.26 to
−88.25 (m), −100.01. HRMS (ESI+) calcd for (C20H24FNO) [M +
H]+ 314.1915, found 314.1918. HPLC purity: 97.58%, retention time:
7.20 min.
(E)-5-((tert-Butoxycarbonyl) (4-(4-Nitrostyryl)phenyl)-
amino)pentyl 4-Methylbenzene-sulfonate (9a). To a 100 mL
round-bottom flask purged with argon and fitted with a magnetic stirrer
were added sodium hydride (0.137 g, 3.43 mmol 60%) and 5-((tert-
butoxycarbonyl)(4-((diethoxyphosphoryl)methyl)phenyl)amino)-
pentyl-4-methylbenzenesulfonate (8, 1.0 g, 1.71 mmol). The mixture
was purged with argon, and dry DMF (4 mL) at 0 °C was added. 4-
Nitrobenzaldehyde (0.260 g, 1.71 mmol) was dissolved in 2 mL of dry
DMF and added dropwise in the reaction mixture after 30 min at 0 °C
under argon. After 3 h, the reaction was monitored with TLC. On
completion, the reaction was quenched with water and the solution was
dissolved in EtOAc and water; the aqueous layer was extracted with
EtOAc (30 mL × 3). The organic layers were combined and washed
with water (50 mL × 2) followed by brine (50 mL). The organic layer
was dried over MgSO4, then filtered and concentrated to yield the
desired product (9a) as a sticky oil. This was purified over a silica flash
(E)-N-(5-Fluoropentyl)-4-(4-methylstyryl)aniline (5f). The
white amorphous ppt. was obtained after water washing and dried
1
under high vacuum, yielding 0.10 g (55%). H NMR (400 MHz,
J
DOI: 10.1021/acs.jmedchem.8b01354
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
column using 15% EtOAc in hexane as eluent, forming a yellow
amorphous compound in a yield of 0.40 g (40%). 1H NMR (400 MHz,
chloroform-d) δ 8.24 (d, J = 8.8 Hz, 2H), 7.78 (dd, J = 8.4, 1.8 Hz, 2H),
7.67−7.61 (m, 2H), 7.55−7.49 (m, 2H), 7.39−7.32 (m, 2H), 7.25−
7.19 (m, 2H), 7.11 (dd, J = 16.3, 6.4 Hz, 1H), 4.05−3.98 (m, 2H), 3.63
(t, J = 7.4 Hz, 2H), 2.47 (s, 3H), 1.70−1.60 (m, 4H), 1.57−1.49 (m,
2H), 1.44 (s, 9H), 1.35 (q, J = 7.8 Hz, 2H). 13C NMR (101 MHz,
chloroform-d) δ 146.9, 145.1, 144.9, 144.0, 132.7, 130.1, 130.0, 128.1,
127.6, 127.5, 127.3, 127.0, 127.0, 126.4, 124.4, 70.5, 70.2, 49.6, 28.7,
28.5, 28.4, 22.8, 21.9, 21.7.
Animal Preparation and Studies. All animal studies were
performed in accordance with a guideline protocol approved by the
Institutional Animal Care and Use Committee of Case Western Reserve
University (Protocol 2016-0023, 2016-0028). The animals were treated
with care to have minimal stress during tail vein injections. Wild-type
C57BL/6 mice (6−8 weeks old, Jackson Laboratory, Bar Harbor, MN)
were used for all in vitro and ex vivo tissue staining, and SD rats (Harlan
Laboratories, Indianapolis, IN) were used for microPET/CT studies.
The rats were fasted overnight prior to imaging but had access to water.
Their diet was then resumed after microPET/CT imaging.
In Vitro Tissue Staining and Assay of Fluorescent Intensity.
Wild-type mice (20−22 g, 8 weeks old) were deeply anesthetized and
perfused transcardially with precooled saline (4 °C, 10 mL/min for 1
min, followed by 7 mL/min for 6 min). This was followed by fixation
with precooled 4% PFA in PBS (4 °C, 10 mL/min for 1 min, followed
by 7 mL/min for 6 min). Brain tissues were then removed, post-fixed by
immersion in 4% PFA overnight, dehydrated in 10, 20, and 30% sucrose
solution, embedded in a freezing compound at optimum cutting
temperature (OCT, Fisher Scientific, Suwanee, GA), and sectioned at
20 μm with a cryostat (Thermo HM525, Thermo Fisher Scientific Inc.,
Chicago, IL). Brain sections were collected from AP (1.0) to AP
(−0.1)33 and were mounted in 12 sections in order at the bottom of 12
superfrost slides (Fisher Scientific) with one section on each slide.
Sections 13−24 were mounted in order in the middle of each slide, and
sections 25−36 were mounted in order on the top of each slide. The
sections were then incubated with test compounds (1 mM, 5% dimethyl
sulfoxide (DMSO) in 1× PBS (pH 7.0), six sections per compound) for
25 min at room temperature in the dark. Excess compounds were
washed by briefly rinsing the slides in PBS (1×) and coverslipped with
Fluoromount-G mounting media (Vector Laboratories, Burlingame,
CA). The sections were then examined under a microscope (Leica
DM4000B, Leica Microsystems Inc., Buffalo Grove, IL) equipped for
fluorescence (DFC7000T), and images of the stained mouse whole
brain sections were taken with the same exposure time.
ImageJ software was then used to quantify pixel intensity values on
six sections of each tested compound. White matter ROIs were selected
on the genu of the corpus callosum (gcc, white matter), and the same-
size ROI was applied to the gray matter on the midline between gcc and
the edge of the section (Figure 4A). The images were analyzed by two
experienced imaging scientists. The FIRs of white matter to gray matter
were then calculated.
Ex Vivo Imaging. Wild-type mice were administered with the
newly synthesized compounds (5a and 5c) (40 mg/kg) via tail vein
injection, and 30 min later, the mice were perfused transcardially with
saline followed by 4% PFA in PBS. Brain tissues were then removed,
post-fixed by immersion in 4% PFA overnight, dehydrated in 30%
sucrose solution, cryostat-sectioned at 100 μm, and mounted on
superfrost slides, and images were acquired directly using a Leica
fluorescence microscope.
Brain Focal Demyelination and in Vitro Staining Treated
with LPC in a Rat Model. Sprague-Dawley female rats (n = 3, 200−
220 g, 6−8 weeks old) were anesthetized by a freshly prepared mixture
of ketamine and xylazine and positioned in a stereotaxic frame
(Stoelting). A small incision was made in the scalp around the region of
the corpus callosum using the following stereotaxic coordinates relative
to the bregma: anterior−posterior, 0.0 mm; medical lateral, 2.0; and
dorsal−ventral, 3.4. A small hole was drilled in the skull and a 26S-gauge
needle attached to a 10 μL Hamilton syringe was lowered into the
corpus callosum according to the dorsal−ventral coordinates. A
microinjector pump (Stoelting) was used to control the infusion of 6
μL lysophosphatidylcholine (LPC, 0.1% in saline) at a rate of 0.25 μL/
min. After infusion, the used needle was left in place for another 2 min
to prevent fluid reflux out of the brain parenchyma. The incision was
then closed via 5-0 Ethicon sutures, and the animals were allowed to
recover on a heating pad. The animals were used for the study after 1
week of recovery. The treated rats were anesthetized and perfused with
saline by 4% PFA. Brain tissues were removed, post-fixed in 4% PFA,
dehydrated in 10, 20, and 30% sucrose solution, and embedded in
OCT. The brain tissues were sectioned at 20 μm intervals with a
cryostat. To determine if the selected test compound can differentiate
demyelinated regions from normal myelinated sheaths, we conducted
in vitro tissue staining using brain sections taken from LPC-treated rats.
LPC-treated brain sections were then incubated with test compound 5a
(1 mM, 5% DMSO in 1× PBS, pH 7.0) for 25 min at room temperature
in the mounting media (Vector Laboratories, Burlingame, CA). The
sections were examined under a microscope (Leica DM4000B, Leica
Microsystems Inc., Buffalo Grove, IL) equipped with fluorescence
(DFC7000T), and images of stained rat whole brain sections were
acquired with the same exposure times. For the comparative study,
standard luxol fast blue (LFB) staining was performed in the meantime
on the adjacent LPC-treated brain tissue sections.
Rat SCI Model. Sprague-Dawley female rats (n = 3, 220−250 g, 8
weeks old) were anesthetized by a freshly prepared mixture of ketamine
and xylazine, and a restricted laminectomy was conducted to expose the
dorsal surface of T13. The vertebral column between T12 and L1 was
then stabilized with clamps and forceps fixed to the base of an Infinite
Horizon Impact Device. The midpoint of T13 was impacted with a
force of 250 kdyn using a 2.5 mm stainless steel impactor tip, which was
used to induce a moderately severe contusive injury to the spinal cord.
The musculature was then sutured over the laminectomy site and the
skin was closed with wound clips, followed by subsequent treatment
with Marcaine at the incision site. The force/displacement graph was
used to monitor impact consistency. After surgery, the animals were
carefully monitored daily for pain and body weight with manual bladder
expression two to three times daily to stimulate reflex voiding until the
animals could urinate independently.
Image Processing for 3D Fluorescence and PET/CT Compar-
ison. To confirm in vivo imaging of spinal cord white matter by PET/
CT, we acquired two-dimensional (2D) fluorescence images and
generated a 3D reconstructed image stack. Fluorescence images were
acquired with a Leica fluorescent microscope and digitized using an in-
built camera with RGB exposure of (R) 210 ms, (G) 2.5 ms, and (B) 1.0
ms. Light from a UV bulb source passed through an excitation filter
(360 nm) and was collected after passing through an emission filter
(400 nm). After extracting the spinal cord and removing the vertebral
bone, sequential cuts were made with 100 μm slice thickness by a
microtome. An image was captured for each slice and numbered
according to the slice number, giving a series of 2D microscopic images
along the length of the spine. The slices were sequentially aligned using
a semiautomated image registration algorithm. Starting from the image
of the first slice, the current slice was used as the reference image for
alignment and the next slice then became the floating image, which was
aligned via control point pairs between corresponding anatomic
features in the two images. A 2D rigid-body transformation (translation,
rotation) was used to transform the floating image to align with the
reference and a minimum of three control point pairs were used for the
transformation. Once two slices were aligned, the algorithm continued
with the newly aligned image serving as the reference image and the
next slice in the sequence served as the floating image. All the slices in
the sequence were aligned by this technique to yield a 3D reconstructed
fluorescence image volume (Figure 10). Image volumes from
fluorescence, PET, and CT were visualized in Amira using volume-
rendering and aligned based on the shape of the spine and fiducials from
the vertebrae (Figure 9).
Radiosynthesis. No carrier-added [18F] sodium fluoride was
produced with a cyclotron (Eclipse High Production, Siemens) via the
nuclear reaction 18O (p,n) 18F. At the end of bombardment, the activity
of aqueous [18F]fluoride ion (50−100 mCi) was transferred to a GE
TRACERlab FXn synthesizer by high-pressure helium.
K
DOI: 10.1021/acs.jmedchem.8b01354
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
The radioactive solution was then passed through a Sep-Pak light
QMA cartridge (Waters, WAT023525) that had been preconditioned
with 5 mL of water, followed by 10 mL of air in a syringe. The
radioactive solution was then eluted with K2CO3 solution (6 mg, 0.043
mmol, in 0.6 mL water), followed by K222 solution (12 mg, 0.032 mmol,
in 1 mL acetonitrile). The solvent was evaporated under a stream of
helium at 85 °C for 5 min, and the residue was subjected to vacuum at
55 °C for another 3 min to get the anhydrous K222/[18F]fluoride ion
complex. A solution of the tosylated precursors (3−5 mg, 0.0062−
0.011 mmol, in 0.8 mL MeCN) was added to this dried complex, and
the mixture was heated at 110 °C for 10 min. EtOAc (3 mL) and hexane
(2 mL) were added to the reaction vessel and the mixture was passed
through a preconditioned Sep-Pak silica cartridge (Waters,
WAT020520; preconditioned with 5 mL of ether). The solvent was
removed under a stream of helium at 70 °C, and the residue was added
to a tin chloride solution (30 mg, 0.16 mmol in 1 mL ethanol and 0.5
mL HCl (1 M)). The resulting mixture was heated at 115 °C for 10−20
min. NaOH solution (0.8 mL, 1 M) and water (15 mL) were then
added, and the resulting mixture was passed through a preconditioned
Sep-Pak C-18 cartridge (Waters, WAT020515 preconditioned with 5
mL of ethanol, followed by 10 mL of water, then dried by 10 mL of air in
a syringe). The cartridge was washed with another 20 mL of water, and
the crude products were eluted with 1 mL of acetonitrile, which was
further purified by semipreparative HPLC (Phenomenex C-18, 10 mm
× 250 mm, MeCN/H2O (60:40), flow rate = 5 mL/min, tR = 9 min).
The radioactive fraction containing the desired products was collected,
diluted with water, loaded onto a Sep-Pak C-18 cartridge, and eluted
with 1 mL of ethanol. After evaporation, the residue was redissolved in
5% ethanol in saline solution and filtered (0.22 μm) into a sterile
injection bottle for animal use. The total time of radiosynthesis was 70
min. Radiochemical purity (RCP) and molar activity (Am) were
determined by analytical HPLC (Phenomenex C-18, 4.6 mm × 250
mm, MeCN/H2O (65:35), flow rate = 1 mL/min, tR = 6−10 min).
Specific activity was calculated by the area of the UV peak of purified F-
18 compound and titrated with the standard curve of the non-
radioactive reference compound of known concentration.
MicroPET/CT Image Acquisition and Analysis. MicroPET/CT
imaging was performed using a Siemens Inveon microPET/CT scanner
in the Case Center for Imaging Research. For better anatomic
localization, CT co-registration was applied. Before microPET imaging,
CT scout views were taken to ensure that the brain tissues were placed
in the co-scan field of view, where the highest image resolution and
sensitivity are achieved. Anesthesia was induced with isoflurane (2−5%
in oxygen). Under anesthesia, radiotracer (1−2 mCi, 37−74 MBq) was
administered via tail vein injection and immediately followed by a
dynamic PET acquisition up to 60 min. Once microPET acquisition
was done, the rat was moved into the CT field and a two-bed CT scan
was performed. A two-dimensional ordered subset expectation
maximization algorithm was used for image reconstruction using CT
for the attenuation correction. For quantitative analysis, the resultant
PET images were registered to the CT images, which enabled us to
accurately define the ROI and quantify the radioactivity concentrations.
In this study, the whole brains of rodents were used as the ROIs and the
radioactivity concentrations were determined in terms of SUV or % ID.
ORCID
Yanming Wang: 0000-0002-2886-2978
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported in part by the National Multiple
Sclerosis Society (RG5174 and PP1703-27324), Genzyme
Corporation, Bryon Riesch Paralysis Foundation, and NIH
(R21 NS111433).
ABBREVIATIONS
■
s, singlet; br s, broad singlet; dd, doublet of doublets; ddd,
doublet of doublet of doublets; br d, broad doublets; t, triplet; dt,
triplet of doublets; q, quartet; m, multiplet; br m, broad
multiplet; NMR, nuclear magnetic resonance; MTR, magnet-
ization transfer ratio; CIC, case imaging compound; BMB,
myelin-binding agent; TAFDAS, triazole fluorinated diamino
stilbene; MeDAS, methyl diamino stilbene; FC2MeDAS, (E)-4-
(4-amino-3-(2-fluoroethoxy)styryl)-N-methylaniline; K222
,
4,7,13,16,21,24-hexaoxa-1,10-diazabicyclo[8.8.8]hexacosane;
NaH, sodium hydride; SnCl2, stannous chloride; K2CO3,
potassium carbonate; MgSO4, magnesium sulfate; HCl, hydro-
chloric acid; THF, tetrahydrofuran; DMF, dimethylformamide;
DCM, dichloromethane; EtOAc, ethyl acetate; NaOH, sodium
hydroxide; TsCl, p-toluenesulfonyl chloride; MeCN, acetoni-
trile; EtOH, ethanol; SUV, standardized uptake value; ID,
injected dose; ROI, region of interest; CT, computerized
tomography; d, deuterated; PFA, paraformaldehyde; RCP,
radiochemical purity; OCT, optimum cutting temperature;
FIR, fluorescence intensity ratio; LPC, lysophosphatidylcholine;
LFB, luxol fast blue; CNS, central nervous system; PET,
positron emission tomography; SMP, small molecular probe;
MRI, magnetic resonance imaging; HPLC, high-performance
liquid chromatography
REFERENCES
■
(1) Ju, G.; Wang, J.; Wang, Y.; Zhao, X. Spinal Cord Contusion.
Neural Regener. Res. 2014, 9, 789−794.
(2) Franklin, R. J. M.; Ffrench-Constant, C. Remyelination in the
CNS: From Biology to Therapy. Nat. Rev. Neurosci. 2008, 9, 839−855.
(3) Mei, F.; Fancy, S. P. J.; Shen, Y. A.; Niu, J.; Zhao, C.; Presley, B.;
Miao, E.; Lee, S.; Mayoral, S. R.; Redmond, S. A.; Etxeberria, A.; Xiao,
L.; Franklin, R. J. M.; Green, A.; Hauser, S. L.; Chan, J. R. Micropillar
Arrays as a High-Throughput Screening Platform for Therapeutics in
Multiple Sclerosis. Nat. Med. 2014, 20, 954−960.
(4) Bai, L.; Lennon, D. P.; Caplan, A. I.; DeChant, A.; Hecker, J.;
Kranso, J.; Zaremba, A.; Miller, R. H. Hepatocyte Growth Factor
Mediates Mesenchymal Stem Cell-Induced Recovery in Multiple
Sclerosis Models. Nat. Neurosci. 2012, 15, 862−870.
(5) Mei, F.; Fancy, S. P.; Shen, Y. A.; Niu, J.; Zhao, C.; Presley, B.;
Miao, E.; Lee, S.; Mayoral, S. R.; Redmond, S. A.; Etxeberria, A.; Xiao,
L.; Franklin, R. J.; Green, A.; Hauser, S. L.; Chan, J. R. Micropillar
Arrays as a High-Throughput Screening Platform for Therapeutics in
Multiple Sclerosis. Nat. Med. 2014, 20, 954−960.
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jmed-
chem.8b01354.
(6) Cruz-Martinez, P.; Gonzalez-Granero, S.; Molina-Navarro, M. M.;
Pacheco-Torres, J.; Garcia-Verdugo, J. M.; Geijo-Barrientos, E.; Jones,
J.; Martinez, S. Intraventricular Injections of Mesenchymal Stem Cells
Activate Endogenous Functional Remyelination in a Chronic
Demyelinating Murine Model. Cell Death Dis. 2016, 7, No. e2223.
(7) Mi, S.; Miller, R. H.; Tang, W.; Lee, X.; Hu, B.; Wu, W.; Zhang, Y.;
Shields, C. B.; Zhang, Y.; Miklasz, S.; Shea, D.; Mason, J.; Franklin, R. J.;
Ji, B.; Shao, Z.; Chedotal, A.; Bernard, F.; Roulois, A.; Xu, J.; Jung, V.;
Pepinsky, B. Promotion of Central Nervous System Remyelination by
NMR, HPLC, and HRMS spectra (PDF)
AUTHOR INFORMATION
■
Corresponding Authors
*E-mail: cxw130@case.edu (C.W.).
*E-mail: yxw91@case.edu (Y.W.).
L
DOI: 10.1021/acs.jmedchem.8b01354
J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Induced Differentiation of Oligodendrocyte Precursor Cells. Ann.
Neurol. 2009, 65, 304−315.
(25) de Paula Faria, D.; Copray, S.; Sijbesma, J. W. A.; Willemsen, A.
T. M.; Buchpiguel, C. A.; Dierckx, R. A. J. O.; de Vries, E. F. PET
Imaging of Focal Demyelination and Remyelination in a Rat Model of
Multiple Sclerosis: Comparison of [11C]Medas, [11C]CIC and
[11C]PIB. Eur. J. Nucl. Med. Mol. Imaging 2014, 41, 995−1003.
(26) Wu, C.; Eck, B.; Zhang, S.; Zhu, J.; Tiwari, A. D.; Zhang, Y.; Zhu,
Y.; Zhang, J.; Wang, B.; Wang, X.; Wang, X.; You, J.; Wang, J.; Guan, Y.;
Liu, X.; Yu, X.; Trapp, B. D.; Miller, R.; Silver, J.; Wilson, D.; Wang, Y.
Discovery of 1,2,3-Triazole Derivatives for Multimodality PET/CT/
Cryoimaging of Myelination in the Central Nervous System. J. Med.
Chem. 2017, 60, 987−999.
(27) Tiwari, A. D.; Wu, C.; Zhu, J.; Zhang, S.; Zhu, J.; Wang, W. R.;
Zhang, J.; Tatsuoka, C.; Matthews, P. M.; Miller, R. H.; Wang, Y.
Design, Synthesis and Evaluation of Fluorinated Radioligands for
Myelin Imaging. J. Med. Chem. 2016, 59, 3705−3718.
(28) Wu, C.; Zhu, J.; Baeslack, J.; Zaremba, A.; Hecker, J.; Kraso, J.;
Matthews, P. M.; Miller, R. H.; Wang, Y. Longitudinal Positron
Emission Tomography Imaging for Monitoring Myelin Repair in the
Spinal Cord. Ann. Neurol. 2013, 74, 688−698.
(29) Demirci, E.; Ahmed, R.; Ocak, M.; Latoche, J.; Radelet, A.;
DeBlasio, N.; Mason, N. S.; Anderson, C. J.; Mountz, J. M. Preclinical
Evaluation of (18)F-ML-10 to Determine Timing of Apoptotic
Response to Chemotherapy in Solid Tumors. Mol. Imaging 2017, 16,
No. 1536012116685941.
(30) Kim, J. W.; Seo, S.; Kim, H. S.; Kim, D. Y.; Lee, H. Y.; Kang, K.
W.; Lee, D. S.; Bom, H. S.; Min, J. J.; Lee, J. S. Comparative Evaluation
of the Algorithms for Parametric Mapping of the Novel Myocardial
PET Imaging Agent (18)F-FPTP. Ann. Nucl. Med. 2017, 31, 469−479.
(31) Rotstein, B. H.; Wey, H. Y.; Shoup, T. M.; Wilson, A. A.; Liang, S.
H.; Hooker, J. M.; Vasdev, N. PET Imaging of Fatty Acid Amide
Hydrolase with [(18)F]DOPP in Nonhuman Primates. Mol. Pharma-
ceutics 2014, 11, 3832−3838.
(32) Parhi, A. K.; Wang, J. L.; Oya, S.; Choi, S.-R.; Kung, M.-P.; Kung,
H. F. 2-(2′-((Dimethylamino)methyl)-4′-(fluoroalkoxy)-phenylthio)-
benzenamine derivatives as Serotonin Transporter Imaging Agents. J.
Med. Chem. 2007, 50, 6673−6684.
(33) Gegrge, P.; Watson, C. The Rat Brain in Stereotaxic Coordinates,
6th ed.; Elsevier Inc: London, U.K., 2007.
(8) Brugarolas, P.; Popko, B. Remyelination Therapy Goes to Trial for
Multiple Sclerosis. Neurol.: Neuroimmunol. Neuroinflammation 2014, 1,
No. e26.
(9) Chen, J. T.; Kuhlmann, T.; Jansen, G. H.; Collins, D. L.; Atkins, H.
L.; Freedman, M. S.; O’Connor, P. W.; Arnold, D. L. Canadian MS/
BMT Group. Voxel-Based Analysis of the Evolution of Magnetization
Transfer Ratio to Quantify Remyelination and Demyelination with
Histopathological Validation in a Multiple Sclerosis Lesion. Neuro-
Image 2007, 36, 1152−1158.
(10) Le Bihan, D.; Mangin, J.-F.; Poupon, C.; Clark, C. A.; Pappata, S.;
Molko, N.; Chabriat, H. Diffusion Tensor Imaging: Concepts and
Applications. J. Magn. Reson. Imaging 2001, 13, 534−546.
(11) Enzinger, C.; Fazekas, F. Measuring Gray Matter and White
Matter Damage in MS: Why This is Not Enough. Front. Neurol. 2015, 6,
No. 56.
(12) Vargas, W. S.; Monohan, E.; Pandya, S.; Raj, A.; Vartanian, T.;
́
Nguyen, T. D.; Hurtado Rua, S. M.; Gauthier, S. A. Measuring
Longitudinal Myelin Water Fraction in New Multiple Sclerosis Lesions.
NeuroImage 2015, 9, 369−375.
(13) Phelps, M. E. Positron Emission Tomography Provides
Molecular Imaging of Biological Processes. Proc. Natl. Acad. Sci.
U.S.A. 2000, 97, 9226−9233.
(14) Stankoff, B.; Wang, Y.; Bottlaender, M.; Aigrot, M. S.; Dolle, F.;
Wu, C.; Feinstein, D.; Huang, G. F.; Semah, F.; Mathis, C. A.; Klunk,
W.; Gould, R. M.; Lubetzki, C.; Zalc, B. Imaging of CNS Myelin by
Positron-Emission Tomography. Proc. Natl. Acad. Sci. U.S.A. 2006, 103,
9304−9309.
(15) Wu, C.; Tian, D.; Feng, Y.; Polak, P.; Wei, J.; Sharp, A.; Stankoff,
B.; Lubetzki, C.; Zalc, B.; Mufson, E. J.; Gould, R. M.; Feinstein, D. L.;
Wang, Y. A Novel Fluorescent Probe That is Brain Permeable and
Selectively Binds to Myelin. J. Histochem. Cytochem. 2006, 54, 997−
1004.
(16) Wang, Y.; Wu, C.; Caprariello, A. V.; Somoza, E.; Zhu, W.; Wang,
C.; Miller, R. H. In Vivo Quantification of Myelin Changes in the
Vertebrate Nervous System. J. Neurosci. 2009, 29, 14663−14669.
(17) Wang, C.; Popescu, D. C.; Wu, C.; Zhu, J.; Macklin, W.; Wang, Y.
In Situ Fluorescence Imaging of Myelination. J. Histochem. Cytochem.
2010, 58, 611−621.
(18) Wu, C.; Wang, C.; Popescu, D. C.; Zhu, W.; Somoza, E. A.; Zhu,
J.; Condie, A. G.; Flask, C. A.; Miller, R. H.; Macklin, W.; Wang, Y. A
Novel PET Marker for In Vivo Quantification of Myelination. Bioorg.
Med. Chem. 2010, 18, 8592−8599.
(19) Frullano, L.; Wang, C.; Miller, R. H.; Wang, Y. A Myelin-Specific
Contrast Agent for Magnetic Resonance Imaging of Myelination. J. Am.
Chem. Soc. 2011, 133, 1611−1613.
(20) Stankoff, B.; Freeman, L.; Aigrot, M. S.; Chardain, A.; Dolle, F.;
Williams, A.; Galanaud, D.; Armand, L.; Lehericy, S.; Lubetzki, C.; Zalc,
B.; Bottlaender, M. Imaging Central Nervous System Myelin by
Positron Emission Tomography in Multiple Sclerosis Using [Methyl-
(1)(1)C]-2-(4′-methylaminophenyl)-6-hydroxybenzothiazole. Ann.
Neurol. 2011, 69, 673−680.
(21) Wang, C.; Wu, C.; Zhu, J.; Miller, R. H.; Wang, Y. Design,
Synthesis, and Evaluation of Coumarin-Based Molecular Probes for
Imaging of Myelination. J. Med. Chem. 2011, 54, 2331−2340.
(22) Condie, A. G.; Gerson, S. L.; Miller, R. H.; Wang, Y. Two-Photon
Fluorescent Imaging of Myelination in the Spinal Cord. ChemMedChem
2012, 7, 2194−2203.
(23) Frullano, L.; Zhu, J.; Miller, R. H.; Wang, Y. Synthesis and
Characterization of a Novel Gadolinium-Based Contrast Agent for
Magnetic Resonance Imaging of Myelination. J. Med. Chem. 2013, 56,
1629−1640.
(24) de Paula Faria, D.; de Vries, E. F. J.; Sijbesma, J. W. A.; Dierckx, R.
A. J. O.; Buchpiguel, C. A.; Copray, S. PET Imaging of Demyelination
and Remyelination in the Cuprizone Mouse Model for Multiple
Sclerosis: A Comparison Between [11C]CIC and [11C]MeDAS.
NeuroImage 2014, 87, 395−402.
M
DOI: 10.1021/acs.jmedchem.8b01354
J. Med. Chem. XXXX, XXX, XXX−XXX