Ligand Specific Efficiency (LSE) Index for PET Tracer Optimization

Article abstract of DOI:10.1002/cmdc.201600112

Ligand efficiency indices are widely used to guide chemical optimization in drug discovery, due to their predictive value in the early steps of optimization. At later stages, however, as more complex properties become critical for success, indices relying on calculated, rather than experimental, parameters become less informative. This problem is particularly acute when developing positron emission tomography (PET) imaging agents, for which nonspecific binding (NSB) to membranes and non-target proteins is a frequent cause of failure. NSB cannot be predicted using in silico parameters. To address this gap, we explored the use of the experimentally determined chromatographic hydrophobicity index on immobilized artificial membranes, CHI(IAM), to guide the optimization of NSB. The ligand specific efficiency (LSE) index was defined as the ratio between affinity (pIC₅₀or pK_d) and the logarithmic value of CHI(IAM). It allows for quantification of binding affinity to the target of interest, relative to NSB. Its use was illustrated by the optimization of PET tracer candidates for the prostacyclin receptor.

Full text of DOI:10.1002/cmdc.201600112

DOI: 10.1002/cmdc.201600112
Full Papers
Ligand Specific Efficiency (LSE) Index for PET Tracer
Optimization
Yves P. Auberson,*[a] Emmanuelle Briard,[a] David Sykes,[b] John Reilly,[c] and Mark Healy[c]
Ligand efficiency indices are widely used to guide chemical op-
timization in drug discovery, due to their predictive value in
the early steps of optimization. At later stages, however, as
more complex properties become critical for success, indices
relying on calculated, rather than experimental, parameters
become less informative. This problem is particularly acute
when developing positron emission tomography (PET) imaging
agents, for which nonspecific binding (NSB) to membranes
and non-target proteins is a frequent cause of failure. NSB
cannot be predicted using in silico parameters. To address this
gap, we explored the use of the experimentally determined
chromatographic hydrophobicity index on immobilized artifi-
cial membranes, CHI(IAM), to guide the optimization of NSB.
The ligand specific efficiency (LSE) index was defined as the
ratio between affinity (pIC50 or pKd) and the logarithmic value
of CHI(IAM). It allows for quantification of binding affinity to
the target of interest, relative to NSB. Its use was illustrated by
the optimization of PET tracer candidates for the prostacyclin
receptor.
Introduction
The use of ligand efficiency (LE) indices in medicinal chemistry
has become widespread over the last decade, providing a con-
venient way to evaluate the quality of a drug candidate at an
early stage. These indices provide information about structural
efficacy with regard to affinity and lipophilicity and about the
enthalpic contribution to binding thermodynamics, in addition
to allowing anticipation of some in vivo properties, such as
oral bioavailability or expected in vivo efficacy.[1] As such, they
have become indispensable tools for medicinal chemists in the
early stages of optimization. Their weakness, unfortunately, is
that with the exception of affinity and lipophilicity measure-
ments, they rely exclusively on calculated properties, such as
the number of non-hydrogen atoms (Nheavy), topological polar
surface area (TPSA), or molecular weight (Mr). This is an advant-
age in early stages of optimization when little experimental
data is available; however, these computational indices increas-
ingly show limitations when approaching more advanced
stages of drug optimization. At this point, candidates differ in
a more subtle manner, where additional parameters come to
the forefront and more complex properties become critical for
success.
Within the field of positron emission tomography (PET)
tracer optimization, multi-parameter optimization (MPO) met-
rics are useful for refining the properties of tracer candidates
for imaging in the central nervous system.[2] Based on pre-
ferred values for clogP, logD at pH 7.4, Mr, TPSA, hydrogen
donor count, and pKa, these MPO metrics allow for initial in sili-
co prioritization of tracer candidates. A limitation, however, is
that these metrics do not take nonspecific binding (NSB) into
account. NSB is the tendency of a molecule to bind indistinctly
to cell membranes, and it is a property that is difficult to pre-
dict. It cannot be quantified by current in silico methods and
remains a challenge in tracer optimization. NSB is clearly one
of the main causes of failure for novel PET tracer candidates.
This property is related in part to lipophilicity, but this parame-
ter does not tell the full story. Therefore, we pursued the de-
velopment of an index that would take measured NSB into ac-
count to guide our efforts to discover clinically successful PET
imaging agents through the whole optimization process.
With this in mind, we decided to explore the use of the ex-
perimentally determined chromatographic hydrophobicity
index on immobilized artificial membranes, CHI(IAM). Our moti-
vation came from the previous observation that the CHI(IAM)
value of PET tracers is statistically correlated to the extent of
their NSB.[3] It is a standardized value derived from the high-
performance liquid chromatography (HPLC) retention time,
which was originally developed to characterize the interactions
of drugs with an immobilized artificial membrane.[4] For practi-
cal reasons, we decided to use CHI(IAM) measurements in
place of the formerly proposed vesicle electrokinetic chroma-
tography (VEKC) assay.[3] CHI(IAM) measurements are easier to
set up, can be automated in higher throughput, and lead to
less variation between individual laboratories. Results obtained
[a] Dr. Y. P. Auberson, Dr. E. Briard
Novartis Institutes for BioMedical Research,
Klybeckstrasse 141, 4057 Basel (Switzerland)
E-mail: yves.auberson@novartis.com
[b] D. Sykes
University of Nottingham, Cell Signalling Department, Nottingham,
NG7 2RD (UK)
[c] Dr. J. Reilly, Dr. M. Healy
Novartis Institutes for BioMedical Research,
250 Massachusetts Avenue, Cambridge MA, 02139 (USA)
The ORCID identification number(s) for the author(s) of this article can
be found under http://dx.doi.org/10.1002/cmdc.201600112.
ChemMedChem 2016, 11, 1 – 14
1
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
with CHI(IAM) are not significantly different from those provid-
ed by the VEKC method.
fallypride for the D2 receptor at 378C at Kd =2 nm.[8] Taking this
lower affinity value into account, the LSE of fallypride would
be 5.8, still within the range of other successful PET imaging
agents. Differences in methods used for the determination of
affinity made it impossible to exactly compare all tracers in
Table 1; however, variations remained within a range that did
not affect the general trend. It is interesting to compare fall-
ypride with the opioid antagonist diprenorphine, which also
has a high affinity for its target (Kd =0.2 nm)[9] but a less favora-
ble CHI(IAM) value of 46.9. This leads to a difference of one
LSE unit and illustrates the importance of NSB on tracer quality.
The two tracers with the lowest LSE values (donepezil: 5.1 and
nomifensine: 5.0) had moderate CHI(IAM) values of 42 and 40,
respectively, but comparatively modest affinities for PET tracers
(IC50 =5.7 nm for AChE,[10] IC50 =10 nm for norepinephrine reup-
take,[11] and Ki =17 nm for dopamine reuptake inhibition[12]).
PK11195, a radioligand well known for its high NSB, has an LSE
value of 5.1. Overall, only two successful PET tracers have LSE
values below 5.4, with an average for all tracers in Table 1 of
5.7. We therefore accepted that a minimum objective to ach-
ieve in PET tracer optimization was LSE>5, preferably LSEꢀ
5.4.
Based on data published by Jiang et al.,[3] CHI(IAM) corre-
lates with the NSB ratio according to the formula CHI(IAM)=
59.4ꢁ24.8ꢁNSB ratio, with a threshold for successful PET trac-
ers of CHI(IAM)ꢂ ~50. The NSB ratio[13] is defined such that
a compound with very low NSB in a brain membrane vs. buffer
dialysis assay has a value close to one, whereas a compound
with high NSB yields a value close to zero. A CHI(IAM)ꢂ ~50
corresponds to a NSB ratio ꢂ0.38, beyond which the specific
signal appears to decrease beyond usefulness.
In analogy to the LE index,[5] which relates the affinity of
a drug to the number of its non-hydrogen atoms, we explored
the usefulness of the ligand specific efficiency (LSE) index,
which we defined as the ratio between affinity (expressed as
the log of its affinity [e.g., pIC50 or pKd] for a specific target),
and the logarithmic value of the experimental nonspecific
binding measurement, CHI(IAM) [Eq. (1)]:
LSE ¼ pKd=logðCHIðIAMÞÞ
ð1Þ
LSE provides a measure of affinity normalized to nonspecific
binding. It shows how efficiently the ligand specifically binds
the desired target, compared with all other nonspecific bind-
ing partners.
An alternative method to determine NSB with a lipid mem-
brane binding assay (LIMBA) has been described recently.[6]
Even though only a small number of tracers were tested in
both assays, a preliminary comparison of LIMBA and CHI(IAM)
indicates that both methods might provide comparable results.
As a consequence, it is conceivable that an index based on
LIMBA logDbrain, as an alternative to CHI(IAM) measurements,
might lead to similar conclusions.
Results and Discussion
Setting the threshold
In a first phase, we evaluated a series of sixteen successful clin-
ical PET tracers to get an understanding of what a desirable
LSE value would be. In all cases, the LSE value was ꢀ5.0
(Table 1), with the dopamine antagonist fallypride reaching
a value of 7.0, due to a very high affinity for the D2 (Kd =
0.03 nm at room temperature) and D3 (Kd =0.3 nm at room
temperature) dopamine receptors,[7] combined with a moderate
propensity for NSB (CHI(IAM)=31.9). Of note, this value is pos-
sibly overestimated, as another study measured the affinity of
At the other end of the range, an NSB ratio of 1 corresponds
to a molecule distributing to the membrane compartment but
having no NSB. This translates into a theoretical CHI(IAM) value
of 34.6. Compounds with CHI(IAM) values significantly below
this number are unlikely to be useful as PET tracers, as this im-
plies that they preferentially distribute outside of the mem-
brane compartment. This might happen, for example, with
very hydrophilic compounds, and is likely to occur in parallel
with low organ penetration. Based on these considerations,
the target CHI(IAM) range for successful PET tracers appears to
be roughly between 35 and 50. This corresponds to the ob-
served CHI(IAM) values of successful tracers (Table 1), all of
which were between 31.9 and 51.9.
Table 1. Affinity for target (pKd), experimental CHI(IAM) values, and calcu-
lated LSE values of sixteen successful clinical PET tracers.
Compound
pKd
CHI(IAM)
LSE
ABP688
AC-5216
8.46
9.52
9.70
8.24
10.52
9.30
9.47
9.60
8.00
9.57
8.35
9.19
8.74
8.75
10.0
9.40
39.1
38.5
46.9
42.2
31.9
42.2
37.4
51.9
40.3
42.9
42.6
48.7
35.4
28.9
47.4
38.9
5.3
6.0
5.8
5.1
7.0
5.7
6.0
5.6
5.0
5.9
5.1
5.4
5.6
6.0
6.0
5.9
Calculating backward from the preferred LSE value of 5.4, it
follows that the target affinity should be at least 0.7 nm for
a CHI(IAM) value of 50 and 4.6 nm for a CHI(IAM) value of 35.
This provides useful guidance in the optimization process, in
the form of a convenient, quantitative relationship between af-
finity and NSB measurements.
diprenorphine
donepezil
fallypride
GR103545
MPPF
NMeSpiperone
nomifensine
NPA
PK11195
PPHT
raclopride
rolipram
SCH23390
WAY100635
A similar analysis to confirm that failed tracers have a lower
LSE value could not be performed, as it was impossible to
tease out the role of NSB among the multiple parameters that
lead to tracer failure. While a certain threshold must be
reached in terms of specific binding for success, the inverse ar-
gument cannot be made, as unsuccessful tracer candidates do
not necessarily fail due to high NSB.
&
ChemMedChem 2016, 11, 1 – 14
www.chemmedchem.org
2
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Target expression can vary greatly (from ~1 to 1000 nm), de-
pending on the target itself, the tissue, and the animal spe-
cies.[14] LSE does not contain any information with regard to
the imaging target, only information regarding the intrinsic
properties of PET tracer candidates. The desirable LSE value is
therefore independent of target expression (Bmax). Clearly,
a tracer for a target expressed at a lower level will require
a higher affinity for the same signal intensity. If all other factors
are proportional, the tracer will therefore benefit from a lower
NSB. The required affinity for a PET tracer has been discussed
elsewhere and as a rule of thumb is given by the formula Kd <
B
max/5.[15] Moreover, the concentration of radiotracer available
for binding to the receptor is dependent on its free fraction,
which can be measured using, for example, tissue homoge-
nates and LC–MS. It also depends on the ratio between specif-
ic and nonspecific binding, a property that can be quantified
and optimized using LSE.
Applying LSE to tracer optimization
The human prostacyclin receptor (hIPR) is a member of the G-
protein-coupled receptor family. Prostacyclin, the major prod-
uct of cyclooxygenase in the macrovascular endothelium, elic-
its potent vasodilation activity and inhibition of platelet aggre-
gation through binding to hIPR. We were interested in evaluat-
ing whether we could develop a PET tracer to study the phar-
macological properties of hIPR ligands, which are of potential
interest for the treatment of pulmonary arterial hypertension
(PAH).[16]
Our chemical starting point was Ro1138452 (1, Figure 1),[17]
which we selected based on encouraging overall properties, in-
cluding a high affinity for hIPR (Ki =0.9 nm, Mr =309, PSA=
45 ꢂ). In contrast, its CHI(IAM) value of 58 and LSE of 4.9 clearly
indicated a high tendency for NSB. This was confirmed in
a binding assay using [3H]1 and membranes of Chinese ham-
ster ovary (CHO) cells expressing hIPR (Figure 1). To better illus-
trate how a higher LSE value translated into better imaging
properties, comparative data is provided for [3H]N-methylsco-
polamine ([3H]NMS). Although not a candidate PET tracer due
to limited brain penetration, [3H]NMS has a very low NSB
(Figure 1) and is a useful tracer for in vitro autoradiography
studies.[18] It has a low CHI(IAM) value of 26.5, which, combined
with a high affinity to muscarinic receptors (pKi =9.2), led to
an LSE value of 6.4. The PET tracer ABP688, which despite
good imaging properties displays some NSB,[19] has an inter-
mediate LSE value of 5.3.
Figure 1. Structure of 1 and saturation binding curves of (upper panel):
[3H]1 binding to hIPR receptors expressed in CHO cell membranes, and for
comparison (lower panel): binding of [3H]NMS to CHO-M3 receptor. Data rep-
resent at least three separate experiments performed in triplicate.
Chemistry
Analogues with fluorinated side chains were obtained in
a manner similar to the original synthesis of 1,[17] starting from
p-substituted phenol 2 (Scheme 1). Introduction of the desired
fluoroalkyl side chain, followed by simultaneous reduction of
the ketone and nitro groups, led to amines 4a–c, which were
transformed into the desired products 5a–c by alkylation with
2-chloro-4,5-dihydro-1H-imidazole.
Scheme 1. Reagents and conditions: a) RX, K2CO3, DMF, 5 h, 708C, quant.;
b) H2, 10% Pd/C, EtOH, HCl, 4.5 h, 508C, 35–91%; c) 2-chloro-4,5-dihydro-1H-
imidazole, iPrOH, 2 h, 858C, 37–90%.
Despite suffering from high NSB, the LSE value of 1 was
close to the minimum value that would be expected for a suc-
cessful PET tracer. This raised hope that some improvement in
binding specificity would allow the use of a close derivative for
imaging purposes. The published structure–activity relation-
ship (SAR) focused on modifications of the substituent on the
left aromatic ring, as well as replacement of the CH2 bridge be-
tween the aromatic moieties and limited variations around the
imidazoline moiety. This left ample space for further structural
modifications to improve the LSE.
Pyridine, pyridazine, and triazine derivatives (Scheme 2) were
obtained from the corresponding methyl heteroaromatic alco-
hols by alkylation with 2-bromopropane to 6a–d, followed by
bromination of the methyl group using N-bromosuccinimide
and AIBN. Compounds 7a–d were then coupled to 4-(4,4,5,5-
tetramethyl-1,3,2-dioxaborolan-2-yl)aniline using a Pd-mediat-
ChemMedChem 2016, 11, 1 – 14
www.chemmedchem.org
3
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Scheme 2. Reagents and conditions: a) For 6a (21%) and 6c (35%): 2-bromo-
propane, K2CO3, NaI, DMF, 808C, 22 h, 21–35%; for 6b (41%) and 6d (47%):
2-bromopropane, Ag2CO3, n-hexane, 858C, 16 h; b) NBS, AIBN, CCl4, 808C,
16 h, 10–90%; c) 4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)aniline,
[Pd(PPh3)4], K3PO4, EtOH/H2O (1:1), DME, microwave, 1508C, 20 min, 49%–
quant.; d) 2-chloro-4,5-dihydro-1H-imidazole, iPrOH, 1 h, 858C, 34–55%.
Scheme 3. Reagents and conditions: a) NaBH4, THF/H2O (1:1), 08C to RT, 4 h,
92%; b) methyl chloroformate, pyridine, THF, 08C to RT, 16 h, 90%; c) allyl-
palladium(II) chloride dimer, dppe, substituted 4-(4,4,5,5-tetramethyl-1,3,2-di-
oxaborolan-2-yl)aniline, K2CO3, DMF, 658C, 15 h, 38–51%; d) 2-chloro-4,5-di-
hydro-1H-imidazole, iPrOH, 2 h, 858C, 35–86%.
ed cross-coupling reaction, and the dihydroimidazole group
was introduced by alkylation, as above, to yield 9a–d in good
overall yields.
The introduction of substituents on the middle aromatic
ring started from 4-isopropoxybenzaldehyde (Scheme 3), which
was reduced to alcohol 10 and transformed into carbonate 11
using methyl chloroformate and pyridine as a base. Pd-mediat-
ed coupling with an appropriately substituted boronate, fol-
lowed by the introduction of the dihydroimidazole, led to
13a–d.
Pyridinones 17a–b (Scheme 4) were prepared starting from
pyridine-2,4-diol, which was selectively alkylated in position 4
with isopropyl bromide. The ring nitrogen was then alkylated
with the corresponding substituted benzyl bromide to yield
15a–b. Reduction of the nitro group and introduction of the
dihydroimidazole as above led to 17a–b in high
yields.
Scheme 4. Reagents and conditions: a) 2-bromopropane, K2CO3, DMF, 658C,
16 h, 48%; b) K2CO3, 1-(bromomethyl)-4-nitrobenzene or 1-(bromomethyl)-2-
fluoro-4-nitrobenzene, THF, 16 h, reflux, 75%; c) Pd/C (10%), NEt3, MeOH, RT,
2 h, 86%; d) 2-chloro-4,5-dihydro-1H-imidazole, iPrOH, 2 h, 858C, 76%.
Compound 23 (Scheme 5), with an inverted pyridi-
none in the left ring, was prepared from 4-bromopyr-
idin-2-ol, first by N-alkylation with 1-bromo-2-methyl-
propane to yield 19, then transformation into the
corresponding pinacolato boronate and Pd-mediated
coupling to 1-(bromomethyl)-4-nitrobenzene to yield
21. Reduction of the nitro group and introduction of
the dihydroimidazole, under the same conditions as
previously used, provided 23.
Scheme 5. Reagents and conditions: a) 1-bromo-2-methylpropane, K2CO3, NaI, DMF, 16 h,
Compound 26 (Scheme 6), with a pyridinone in 608C, 61%; b) bis(pinacolato)diboron, PdCl2(dppf)·CH2Cl2, KOAc, dioxane, microwave,
1208C, 1 h, 95%; c) 1-(bromomethyl)-4-nitrobenzene, [PdCl2(dppf)·CH2Cl2], K2CO3, DME,
EtOH/H2O, microwave, 1208C, 10 min, 34%; d) Pd/C (10%), H2, NEt3, EtOH, RT, 2 h, 90%;
e) 2-chloro-4,5-dihydro-1H-imidazole, iPrOH, 40 min, 858C, 39%.
the center of the molecule, was prepared by bromi-
nation of (4-isopropoxyphenyl)methanol, followed by
alkylation of 4-chloropyridin-2(1H)-one to yield 25.
The dihydroimidazolamine group was introduced by
Pd-mediated coupling and yielded the desired prod-
uct in low yield. No attempt was made to further op-
timize this step.
Finally, replacement of the dihydroimidazolamine
group was achieved by the following procedures
(Scheme 7). A dihydrothiazolamine was introduced
Scheme 6. Reagents and conditions: a) CBr4, PPh3, CH2Cl2, RT, 1 h, 42%; b) 4-chloropyridin-
2(1H)-one, K2CO3, THF, microwave, 1308C, 30 min; c) 4,5-dihydro-1H-imidazol-2-amine,
by treating 12b (R1 =iPr, R2 =F) and 27a[20] (R1 =iPr,
R2 =H) with benzoyl isothiocyanate, followed by alka- [BrettPhos-PdII], BrettPhos, LiHMDS (1m in THF), THF, microwave, 1308C, 30 min, 7%.
&
ChemMedChem 2016, 11, 1 – 14
www.chemmedchem.org
4
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
ortho or meta position on the ring adjacent to the
imidazoline decreased it slightly, to 9.6 (13a) and 9.8
(13b), with no effect on LSE. Other substituents
(13c–d) did not have a stronger influence on basicity.
While unsuccessful at influencing this property, these
derivatives provided interesting SAR information. In
particular, the introduction of a fluorine atom in the
meta position (13b) was well-tolerated and led to
a slight increase in affinity for hIPR.
To further explore the potential of increasing LSE
by introducing polarity in the left part of the mole-
cule, we prepared pyridinone derivatives 17a and 23.
Both showed a clear improvement in CHI(IAM) values
but lost affinity for hIPR. Compound 17a, neverthe-
less, retained an LSE value of 4.9. Encouraged by this
result, we tried to regain potency by introducing a flu-
orine atom on the left ring (17b), in analogy to 13b.
To our surprise, the affinity for hIPR actually de-
creased, while a concomitant increase in NSB led to
an LSE value of only 3.5.
Scheme 7. Reagents and conditions: a) benzoyl isothiocyanate, acetone, RT, 1 h, 77%;
b) aq NaOH (5%), 908C, 20 min, quant.; c) 2-bromoethylamine, EtOH/H2O (8:1), reflux,
16 h, 32%; d) 1-chloro-2-isocyanatoethane, CH2Cl2, RT, 16 h, quant.; e) SiO2·KF, CH3CN,
reflux, 4 h, 70%; f) 5-chloro-3,4-dihydro-2H-pyrrole, toluene, reflux, 2 h, 55%; g) 6-chloro-
2,3,4,5-tetrahydropyridine, toluene, reflux, 2 h, 49–60%.
A first hint that the imidazoline group could be re-
placed while retaining some affinity was obtained
with 29a. In contrast, the CHI(IAM) value of dihy-
drooxazole analogue 31 did not improve enough to
line hydrolysis and cyclization with 2-bromoethylamine to yield
Table 2. Affinity for hIPR (pKi), experimental CHI(IAM) values, and calculat-
ed LSE values of hIPR ligands.
29a–b. The dihydrooxazolamine derivative 31 was obtained by
treating 4-(4-isopropoxybenzyl)aniline[17] with 1-chloro-2-isocya-
natoethane, followed by cyclization with KF, in the presence of
SiO2 as a phase-transfer catalyst. The fluorinated anilines 4a
(R1 =nBuF, R2 =F) and 4b (R1 =nPrF, R2 =F) were alkylated with
five- and six-membered imidoyl chlorides to yield the dihy-
droaminopyrrole and tetrahydroaminopyridine derivatives
32a–b and 33a–b.
Compound
pKi
CHI(IAM)
LSE
1
5a
5b
5c
9a
9b
9c
9d
13a
13b
13c
13d
17a
17b
23
8.68
9.09
8.75
7.56
7.33
7.91
5.10
7.87
8.03
9.28
7.27
8.33
7.06
5.92
5.88
4.93
7.48
7.66
6.78
8.44
8.46
8.06
8.51
57.7
56.8
55.0
50.0
56.6
54.0
45.1
49.4
56.7
60.0
64.6
56.3
28.0
36.3
33.7
31.2
47.8
48.1
39.8
49.7
46.6
49.2
50.0
4.9
5.2
5.0
4.4
4.2
4.6
3.1
4.6
4.6
5.2
4.0
4.8
4.9
3.8
3.8
3.3
4.5
4.6
4.2
5.0
5.0
4.8
5.0
Optimization and pharmacological evaluation
Before engaging in a broader exploration program, we first en-
sured that a fluorine atom could be introduced on the oxyalkyl
side chain of 1 without deleterious effects. We selected side
chains known to be convenient for preparation of precursors
and later radiolabeling by classical nucleophilic substitution
using 18Fꢁ (5a–c, Table 2). Encouragingly, the first two varia-
tions had no effect on LSE, while 5c appeared less attractive,
as the decrease in its CHI(IAM) value did not compensate for
its lower affinity for the receptor.
26
29a
29b
31
32a
32b
33a
33b
We then tried to introduce polarity in the left aromatic ring,
a measure which effectively spread polarity across the mole-
cule and was expected to improve LSE. Compounds 9a–b, the
pyridine analogues of 1, however, showed no significant im-
provement. Introducing two nitrogen atoms into the ring (9c–
d) led to a marked decrease in CHI(IAM) values but was paral-
leled by a strong reduction in affinity and did not improve LSE.
Suspecting that the imidazoline might play an important
role in driving not only affinity but also NSB, we then tried to
influence the latter by modulating its basicity. The pKb value of
1 in its protonated form was 10.5; introducing a fluorine in the
compensate for the compound’s loss in affinity. Compound
29b, the fluoro analogue of 29a, was only slight improved. In
contrast, 32a and 33a regained potency, and thanks to lower
CHI(IAM) values, again reached LSE values equivalent to those
of compound 1.
Despite synthesizing a number of additional derivatives, we
struggled to identify compounds with significantly higher LSE
ChemMedChem 2016, 11, 1 – 14
www.chemmedchem.org
5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
8.7 Hz, 2H), 7.81 (d, J=8.7 Hz, 2H), 7.74 (d, J=8.8 Hz, 2H), 6.91 (d,
J=8.8 Hz, 2H), 4.8 (dt, J=48 Hz, J=5.6 Hz, 2H), 4.05 (t, J=6.0 Hz,
2H), 1.76–1.96 ppm (m, 4H); UPLC–MS: tR =1.2 min, m/z: 318.0
[M+H].
values in this series. We hence decided to prepare fluoroalkyl
analogues of the best candidates to assess their potential, in-
cluding compounds 32b and 33b. The affinity and CHI(IAM)
values of these compounds did not differ significantly, leading
to identical LSE values and showing only marginal improve-
ment over the lead molecule. We therefore concluded that this
series did not have enough potential for the development of
PET imaging agents. To confirm this result, we nevertheless de-
cided to tritiate 33b and test it in an autoradiography study
looking at lung tissue in the presence and absence of a blocker.
We expected to see at least a modest signal if 33b had prop-
erties deserving further evaluation. The images obtained in
this experiment (not shown) did not display any significant dif-
ferences in signal intensity with or without blocking, confirm-
ing that 33b did not have potential for development as a PET
imaging agent.
4-(4-(4-Fluorobutoxy)benzyl)aniline (4a): Concentrated aq HCl
(0.22 mL, 7.40 mmol) was added to a suspension of 3a (150 mg,
0.47 mmol) and Pd/C (15 mg, 0.014 mmol) in EtOH (2.1 mL). The re-
action mixture was stirred under hydrogen (3.5 bar) for 4.5 h at
508C, cooled to room temperature, and filtrated through Celite.
The Celite cake was washed with EtOH, and the filtrate was con-
centrated and dissolved in a mixture of EtOAc and saturated aq
NaHCO3. The aqueous layer was extracted twice with EtOAc, and
the combined organic layers were washed with brine, dried over
anhydrous Na2SO4, filtered, and concentrated to give 4a (117 mg,
91%) as a white solid: 1H NMR (400 MHz, CDCl3): d=7.00 (d, J=
8.56 Hz, 2H), 6.89 (d, J=8.19 Hz, 2H), 6.73 (d, J=8.0 Hz, 2H), 6.55
(d, J=8.0 Hz, 2H), 4.50 (dt, J=48 Hz, J=5.55 Hz, 2H), 3.90 (t, J=
5.75 Hz, 2H), 1.66–1.86 ppm (m, 4H); UPLC–MS: tR =1.08 min, m/z:
274.1 [M+H].
Conclusions
N-(4-(4-(4-Fluorobutoxy)benzyl)phenyl)-4,5-dihydro-1H-imidazol-
2-amine (5a): A suspension of 4a (117 mg, 0.41 mmol) and 2-
chloro-4,5-dihydro-1H-imidazole (109 mg, 0.54 mmol) in isopropa-
nol (1.5 mL) was stirred at 80–858C in a sealed glass tube for 2 h.
The reaction mixture was cooled to room temperature, diluted in
EtOAc, and washed with diluted aq NaHCO3. The aqueous layer
was extracted twice with EtOAc. The combined organic layer was
washed with brine, dried over anhyd Na2SO4, filtered, and concen-
trated. The crude product was purified on silica gel by flash chro-
matography (CH2Cl2/MeOH:NH3 [9:1], from 100:0 to 90:10), to give
The use of LSE as an index of suitability for PET imaging is
based on a series of clinically validated tracers and is an easily
applied tool that proved useful in this and other projects.
Clearly, its application to a diverse series of PET tracer candi-
dates will be needed to definitively confirm its predictive
value. Nevertheless, LSE is based on a rather intuitive concept,
in the sense that a good PET tracer candidate should have
a balanced mix of affinity and binding specificity. It is a conven-
ient index to evaluate and compare molecules based on mea-
sured, rather than in silico, values and is applicable independ-
ently of target and chemotype. It can be used to compare di-
verse starting points for optimization programs, as well as for
candidate prioritization and selection.
1
5a (126 mg, 90%) as a white solid: H NMR (400 MHz, CDCl3): d=
7.05–6.97 (m, 4H) 6.84 (d, J=8.31 Hz, 2H) 6.74 (d, J=7.67 Hz, 2H)
4.45 (dt, J=48 Hz, J=5.6 Hz, 2H) 3.91 (t, J=5.81 Hz, 2H) 3.79 (s,
2H) 3.45 (s, 4H) 2.22–1.90 (brs, 2H) 1.76–1.89 ppm (m, 4H); UPLC–
MS: tR =0.84 min, m/z: 342.4 [M+H].
(4-(3-Fluoropropoxy)phenyl)(4-nitrophenyl)methanone
(3b):
Compound 3b was prepared from 2 (108 mg, 0.44 mmol), accord-
ing to the procedure described for the synthesis of 3a. After purifi-
cation by flash chromatography, 3b was obtained as a white
Experimental Section
1
powder (144 mg, quant.): H NMR (400 MHz, CDCl3): d=8.35 (d, J=
Chemistry
8.68 Hz, 2H), 7.89 (d, J=8.68 Hz, 2H), 7.81 (d, J=8.80 Hz, 2H), 6.99
(d, J=8.80 Hz, 2H), 4.55 (dt, J=44 Hz, 5.6 Hz, 2H), 4.12 (t, J=
6.05 Hz, 2H), 1.86–2.02 ppm (m, 2H); UPLC–MS: tR =1.20 min, m/z:
318.0 [M+H].
General methods: All chemicals, reagents, and solvents were of
analytical grade, purchased from commercial sources, and used
without purification, unless otherwise specified. 1H NMR spectra
were acquired on a Bruker (400 MHz) or Bruker Advance (600 MHz)
spectrophotometer. d values are given in parts per million (ppm)
relative to the residual solvent peak. Analytical LCMS/HPLC condi-
tions (%=percent by volume): UPLC-ZQ2000, Acquity HSS-T3 (2.1ꢁ
50 mm, 1.8 mm) column; room temperature; mobile phase:
water+0.5–1.0% HCO2H (A)/5% acetonitrile+0.5–1.0% HCO2H (B);
gradient: from 2% to 98% B in 4.3 min+0.7 min isocratic; flow
4-(4-(3-Fluoropropoxy)benzyl)aniline (4b): Compound 4b was
prepared from 3b (229 mg, 0.38 mmol), according to the proce-
dure described for the synthesis of 4a, and was obtained as
a white powder (81 mg, 83%): UPLC–MS: tR =1.01 min, m/z: 260.3
[M+H].
N-(4-(4-(3-Fluoropropoxy)benzyl)phenyl)-4,5-dihydro-1H-imida-
zol-2-amine (5b): Compound 5b was prepared from 4b (81 mg,
0.31 mmol), according to the procedure described for the synthesis
rate: 1.0 mLminꢁ1
.
(4-(4-Fluorobutoxy)phenyl)(4-nitrophenyl)methanone (3a):
A
1
of 5a, and was obtained as a white powder (39 mg, 37%): H NMR
suspension of 2[17] (108 mg, 0.44 mmol), 1-bromo-4-fluorobutane
(195 mL, 1.78 mmol), K2CO3 (92 mg, 0.67 mmol), and NaI (3.7 mg,
0.02 mmol) in DMF (600 mL) was stirred at 65–708C in a sealed
glass tube for 5 h. The reaction mixture was diluted in EtOAc and
washed with water. The aqueous layer was extracted twice with
EtOAc. The combined organic layers were washed with brine, dried
over anhydrous Na2SO4, filtered, and concentrated. The crude prod-
uct was purified by flash chromatography on silica gel (cyclohex-
ane/EtOAc, 100:0 to 50:50) to give 3a (144 mg, quant.) as a white
(400 MHz, CDCl3): d=7.09 (dd, J=19.75, 8.38 Hz, 4H), 6.90 (d, J=
8.31 Hz, 2H), 6.83 (d, J=8.56 Hz, 2H), 4.65 (dt, J=44 Hz, J=
5.81 Hz, 2H), 4.08 (t, J=6.11 Hz, 2H), 3.86 (s, 2H), 3.51 (s, 4H),
2.09–2.23 ppm (m, 2H); 13C NMR (151 MHz, CDCl3): d=157.8, 157.1,
147.5, 135.2, 134.0, 129.9 (2C), 129.6 (2C), 122.9, 114.4 (2C), 81.4,
80.3, 63.5, 42.5, 40.5 (2C), 30.5 ppm; UPLC–MS: tR =0.73 min, m/z:
328.1 [M+H].
1
solid: H NMR (400 MHz, CDCl3): d=8.26–8.22 (m, 2H), 8.27 (d, J=
&
ChemMedChem 2016, 11, 1 – 14
www.chemmedchem.org
6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
(4-(2-(2-Fluoroethoxy)ethoxy)phenyl)(4-nitrophenyl)methanone
4.17(m, 2H), 3.83–3.92 (m, 5H), 3.76–3.81 (m, 1H), 3.47–3.54 ppm
(3c): TBAF (3.94 mL, 3.94 mmol) was added to a solution of 2-(2-(4-
(m, 4H); UPLC–MS: tR =0.68 min, m/z: 358.4 [M+H].
(4-nitrobenzoyl)phenoxy)ethoxy)ethyl
4-methylbenzenesulfonate
5-Isopropoxy-2-methylpyridine (6a): A suspension of 6-methyl-
pyridin-3-ol (1.0 g, 9.16 mmol), 2-bromopropane (2.6 mL,
27.5 mmol), K2CO3 (1.9 g, 13.75 mmol), and NaI (0.076 g,
0.50 mmol) in DMF (10.00 mL) was stirred in a sealed glass tube at
808C for 22 h under argon. The reaction mixture was diluted with
EtOAc, washed with water and brine, dried over a phase separator,
and concentrated. The residue was purified on silica gel by flash
chromatography (cyclohexane/EtOAc, 100:0 to 0:100) to afford 6a
(239 mg, 0.49 mmol) in THF (4.10 mL), and the resulting deep
purple solution was stirred at 658C for 2 h in a sealed glass tube
under nitrogen. The solvent was then evaporated. The residue was
dissolved in EtOAc, and the organic layer was washed with water.
The aqueous layer was extracted twice with EtOAc. The combined
organic layers were washed with brine, dried over anhyd Na2SO4,
filtered, and concentrated. The crude was purified on silica gel by
flash chromatography (cyclohexane/EtOAc, 100:0 to 85:15) to give
(288 mg, 20.8%) as
a
colorless liquid: 1H NMR (400 MHz,
1
3c (103.8 mg, 63.3%) as a yellow liquid: H NMR (400 MHz, CDCl3):
[D6]DMSO): d=8.10 (d, J=2.93 Hz, 1H), 7.26 (dd, J=8.50, 3.00 Hz,
1H), 7.15 (d, J=8.56 Hz, 1H), 4.62 (dt, J=12.07, 6.01 Hz, 1H), 2.38
(s, 3H), 1.26 ppm (d, J=5.99 Hz, 6H).
d=8.34 (d, J=8.80 Hz, 2H), 7.89 (d, J=8.68 Hz, 2H), 7.81 (d, J=
8.7 Hz, 2H), 7.03 (d, J=8.7 Hz, 2H), 4.65–4.70 (m, 1H), 4.49–4.62
(m, 1H), 4.22–4.28 (m, 2H), 3.92–3.99 (m, 2H), 3.86–3.91 (m, 1H),
3.76–3.85 ppm (m, 1H); UPLC–MS: tR =1.07 min, m/z: 334.0 [M+
H].
2-(Bromomethyl)-5-isopropoxypyridine (7a): NBS (339 mg,
1.90 mmol) and AIBN (31.3 mg, 0.19 mmol) were added to a solu-
tion of 6a (288 mg, 1.90 mmol) in CCl4 (6.35 mL) at room tempera-
ture under argon. The reaction mixture was stirred at 808C for
60 h. The reaction mixture was then cooled to room temperature
and extracted with CH2Cl2 and water. The organic layer was dried
over a phase separator and concentrated. The crude product was
purified on silica gel by flash chromatography (cyclohexane/EtOAc,
100:0 to 70:30) to afford 7a (85 mg, 9.7%) as an orange oil with
50% purity. The product was used as such in the next step.
The precursor was prepared in the following manner:
2-(2-(4-(4-Nitrobenzoyl)phenoxy)ethoxy)ethyl 4-methylbenzene-
sulfonate: p-Tolylsulfonyl chloride (133 mg, 0.69 mmol) followed
by triethylamine (230 mL, 1.65 mmol) were added successively at
0–58C under nitrogen to a stirred solution of (4-(2-(2-hydroxye-
thoxy)ethoxy)phenyl)(4-nitrophenyl)methanone
(218.6 mg,
0.66 mmol) in CH2Cl2 (528 mL). The resulting solution was stirred at
room temperature for 18 h, then diluted with CH2Cl2. The organic
layer was washed with water, brine, dried over anhyd Na2SO4, fil-
tered and concentrated. The crude product was purified by flash
chromatography on silica gel (cyclohexane/EtOAc 100:0 to 45:55)
to give the desired product (239 mg, 72.5%) as transparent liquid.
1H NMR (400 MHz, CDCl3): d=8.35 (d, J=8.68 Hz, 2H), 7.89 (d, J=
8.68 Hz, 2H), 7.73–7.85 (m, 4H), 7.33 (d, J=8.31 Hz, 2H), 7.00 (d,
J=8.80 Hz, 2H), 4.19 (dt, J=12.26, 4.69 Hz, 4H), 3.82–3.89 (m, 2H),
3.75–3.81 (m, 2H), 2.44 ppm (s, 3H); UPLC–MS: tR =1.21 min, m/z:
486.0 [M+H].
4-((5-Isopropoxypyridin-2-yl)methyl)aniline (8a): DME (1.7 mL)
and EtOH/H2O (0.4 mL each) were added to a mixture of 7a
(85 mg, 0.18 mmol), 4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)a-
niline (50.1 mg, 0.22 mmol), [Pd(PPh3)4] (21.34 mg, 0.018 mmol),
and K3PO4 (78 mg, 0.369 mmol). Argon was bubbled through the
stirred reaction mixture for 5 min, then the mixture was heated in
a microwave at 1208C for 40 min, cooled to room temperature,
concentrated, and purified on silica gel by flash chromatography
(cyclohexane/EtOAc, 100:0 to 50:50) to afford 8a as a yellow oil
1
(32 mg, 49%): H NMR (400 MHz, [D6]DMSO): d=8.01–8.36 (m, 1H);
(4-(2-(2-Hydroxyethoxy)ethoxy)phenyl)(4-nitrophenyl)metha-
none: A suspension of 2 (150 mg, 0.62 mmol), 2-(2-chloroethoxy)-
ethan-1-ol (267 mL, 2.47 mmol), K2CO3 (128 mg, 0.92 mmol) and NaI
(5.2 mg, 0.035 mmol) in DMF (833 mL) was stirred at 65–708C in
a sealed glass tube for 16 h. After cooling to room temperature,
the reaction mixture was diluted in EtOAc and washed with water.
The aqueous layer was extracted twice with EtOAc. The combined
organic layers were washed with brine, dried over anhyd Na2SO4,
filtered and concentrated. The crude was purified on silica gel by
flash chromatography (cyclohexane/EtOAc, 100:0 to 20:80) to pro-
vide the desired product as a white solid (218.6 mg, 100%).
1H NMR (400 MHz, CDCl3): d=8.34 (d, J=8.80 Hz, 2H), 7.89 (d, J=
8.80 Hz, 2H), 7.82 (d, J=8.80 Hz, 2H), 7.02 (d, J=8.80 Hz, 2H), 4.18
ꢁ4.29 (m, 2H), 3.86–3.97 (m, 2H), 3.76–3.83 (m, 2H), 3.66–3.73 (m,
2H), 1.99–2.05 ppm (m, 1H); UPLC–MS: tR =0.90 min, m/z: 332.1
[M+H].
7.41–7.76 (m, 3H), 6.98–7.34 (m, 2H), 6.87 (dd, J=8.19, 2.32 Hz,
1H), 6.11–6.67 (m, 2H), 4.85 (brs, 1H), 4.60 (ddt, J=8.94, 6.01, 3.16,
3.16 Hz, 1H), 3.79 (d, J=2.20 Hz, 1H), 1.24 ppm (dd, J=5.99,
2.69 Hz, 6H); UPLC–MS: tR =0.70 min, m/z: 243.4 [M+H].
N-(4-((5-Isopropoxypyridin-2-yl)methyl)phenyl)-4,5-dihydro-1H-
imidazol-2-amine (9a): Compound 8a (32 mg, 0.13 mmol) and 2-
chloro-4,5-dihydro-1H-imidazole (37.2 mg, 0.17 mmol) in isopropa-
nol (1.8 mL) were stirred at 858C in a sealed glass tube for 1 h
under argon. The reaction mixture was cooled to room tempera-
ture and concentrated, and the residue was extracted with CH2Cl2
and 2m aq NaOH. The organic layer was washed with brine, dried
over a phase separator, and concentrated. The crude product was
purified on silica gel by flash chromatography ([CH2Cl2:MeOH]/10%
NH4OH, 100:0 to 80:20) to afford 9a as an orange oil (17 mg,
33.5%): 1H NMR (400 MHz, CD3OD): d=8.08 (d, J=2.81 Hz, 1H),
7.33 (dd, J=8.68, 2.93 Hz, 1H), 7.13–7.27 (m, 3H), 6.94–7.06 (m,
2H), 4.64 (dt, J=12.07, 6.01 Hz, 1H), 4.03 (s, 2H), 3.53 (d, J=
2.57 Hz, 4H), 1.34 ppm (d, J=5.99 Hz, 6H); UPLC–MS: tR =0.61 min,
m/z: 311.4 [M+H].
4-(4-(2-(2-Fluoroethoxy)ethoxy)benzyl)aniline (4c): Compound
4c was prepared from 3c (83 mg, 0.25 mmol), according to the
procedure described for the synthesis of 4a, and was obtained as
a white oil (25.6 mg, 35%): UPLC–MS: tR =0.91 min, m/z: 290.1
[M+H].
2-Isopropoxy-5-methylpyridine (6b): 2-bromopropane (2.58 mL,
27.5 mmol), Ag2CO3 (3.79 g, 13.75 mmol), and NaI (0.069 g,
0.46 mmol) were added to 5-methylpyridin-2-ol (1.00 g, 9.16 mmol)
in n-hexane (40 mL) at 858C under argon. After cooling to room
temperature, the reaction mixture was stirred in a sealed glass
tube at 858C for 16 h. The insoluble residue was filtered through
Celite, the Celite was washed with n-hexane, and the filtrate was
concentrated and purified on silica gel by flash chromatography
N-(4-(4-(2-(2-Fluoroethoxy)ethoxy)benzyl)phenyl)-4,5-dihydro-
1H-imidazol-2-amine (5c): Compound 5c was prepared from 4c
(25 mg, 0.0.9 mmol), according to the procedure described for the
synthesis of 5a, and was obtained as a colorless oil (14 mg,
42.7%): 1H NMR (400 MHz, CDCl3): d=7.09 (dd, J=17.91, 8.38 Hz,
4H), 6.79–6.95 (m, 4H), 4.62–4.69 (m, 1H), 4.49–4.56 (m, 1H), 4.08–
ChemMedChem 2016, 11, 1 – 14
www.chemmedchem.org
7
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
(cyclohexane/EtOAc, 100:0 to 95:5) to provide 6b as a colorless
liquid (630 mg, 41%): H NMR (400 MHz, [D6]DMSO): d=7.94 (d, J=
2.08 Hz, 1H), 7.49 (dd, J=8.31, 2.45 Hz, 1H), 6.63 (d, J=8.44 Hz,
1H), 5.18 (dt, J=12.35, 6.17 Hz, 1H), 2.19 (s, 3H), 1.26 ppm (d, J=
6.11 Hz, 6H); UPLC–MS: tR =1.03 min, m/z: 152.1 [M+H].
2-Isopropoxy-5-methylpyrazine (6d): Compound 6d was pre-
pared from 5-methylpyrazin-2-ol (45 mg, 0.19 mmol), according to
the procedure described for the synthesis of 6b, and was obtained
as an orange liquid (395 mg, 46.9%). It was used in the next step
without further purification.
1
5-(Bromomethyl)-2-isopropoxypyridine (7b): Compound 7b was
prepared from 6b (464 mg, 3.07 mmol), according to the proce-
dure described for the synthesis of 7a, and was obtained as a color-
less oil (159 mg, 13.5%). The product was not stable and was used
as such in the next step.
2-(Bromomethyl)-5-isopropoxypyrazine (7d): Compound 7d was
prepared from 6d (489 mg, 3.21 mmol), according to the proce-
dure described for the synthesis of 7a, and was obtained as an
orange oil (338 mg, 68%): 1H NMR (400 MHz, [D6]DMSO): d=8.34
(d, J=0.98 Hz, 1H), 8.23 (d, J=1.10 Hz, 1H), 5.23 (dt, J=12.35,
6.17 Hz, 1H), 4.73 (s, 2H), 1.25–1.36 ppm (m, 6H); UPLC–MS: tR =
1.04 min, m/z: 232.9 [M+H].
4-((6-Isopropoxypyridin-3-yl)methyl)aniline (8b): Compound 8b
was prepared from 7b (159 mg, 0.43 mmol), according to the pro-
cedure described for the synthesis of 8a, and was obtained as
4-((5-Isopropoxypyrazin-2-yl)methyl)aniline (8d): Compound 8d
was prepared from 7d (85 mg, 0.34 mmol), according to the proce-
dure described for the synthesis of 8a, and was obtained as an
orange oil (81 mg, 72%): 1H NMR (400 MHz, [D6]DMSO): d=8.09–
8.14 (m, 1H), 8.04 (s, 1H), 6.89 (d, J=8.31 Hz, 2H), 6.47 (d, J=
8.31 Hz, 2H), 5.07–5.25 (m, 1H), 3.82 (s, 2H), 1.29 ppm (d, J=
6.24 Hz, 6H); UPLC–MS: tR =0.89 min, m/z: 244.1 [M+H].
1
a yellow oil (13 mg, 12.6%): H NMR (400 MHz, [D6]DMSO): d=7.97
(d, J=2.20 Hz, 1H), 7.43 (dd, J=8.44, 2.45 Hz, 1H), 6.85 (d, J=
8.19 Hz, 2H), 6.62 (d, J=8.44 Hz, 1H), 6.47 (d, J=8.31 Hz, 2H), 5.18
(dt, J=12.35, 6.17 Hz, 1H), 4.87 (s, 2H), 3.65 (s, 2H), 1.25 ppm (d,
J=6.11 Hz, 6H); UPLC–MS: tR =0.98 min, m/z: 243.1 [M+H].
N-(4-((6-Isopropoxypyridin-3-yl)methyl)phenyl)-4,5-dihydro-1H-
imidazol-2-amine (9b): Compound 9b was prepared from 8b
(108 mg, 0.44 mmol), according to the procedure described for the
synthesis of 9a, and was obtained as a white oil (14 mg, 54.6%):
1H NMR (400 MHz, CD3OD): d=7.97 (d, J=2.20 Hz, 1H), 7.50 (dd,
J=8.56, 2.45 Hz, 1H), 7.17 (d, J=8.19 Hz, 2H), 7.01 (d, J=8.44 Hz,
2H), 6.66 (d, J=8.56 Hz, 1H), 5.16 (dt, J=12.35, 6.17 Hz, 1H), 3.87
(s, 2H), 3.55 (s, 4H), 1.32 ppm (d, J=6.11 Hz, 6H); UPLC–MS: tR =
0.75 min, m/z: 311.4 [M+H].
N-(4-((5-Isopropoxypyrazin-2-yl)methyl)phenyl)-4,5-dihydro-1H-
imidazol-2-amine (9d): Compound 9d was prepared from 8d
(81 mg, 0.25 mmol), according to the procedure described for the
synthesis of 9a, and was obtained as a yellow oil (40 mg, 50.4%):
1H NMR (400 MHz, CD3OD): d=8.03 (d, J=9.90 Hz, 2H), 7.19 (d, J=
8.31 Hz, 2H), 6.93–7.06 (m, 2H), 5.26 (dt, J=12.35, 6.17 Hz, 1H),
4.03 (s, 2H), 3.52 (s, 4H), 1.35 ppm (d, J=6.11 Hz, 6H); UPLC–MS:
tR =0.71 min, m/z: 312.4 [M+H].
3-Isopropoxy-6-methylpyridazine (6c): Compound 6c was pre-
pared from 6-methylpyridazin-3-ol (1 g, 9.08 mmol), according to
the procedure described for the synthesis of 6a, and was obtained
as an orange oil (489 mg, 35.4%): 1H NMR (400 MHz, [D6]DMSO):
d=7.29 (d, J=9.41 Hz, 1H), 6.83 (d, J=9.54 Hz, 1H), 5.11 (quin, J=
6.66 Hz, 1H), 2.28 (s, 3H), 1.25 ppm (d, J=6.72 Hz, 6H); UPLC–MS:
tR =0.68 min, m/z: 153.0 [M+H].
(4-Isopropoxyphenyl)methanol (10): A suspension of 4-isopropox-
ybenzaldehyde (1.0 g, 6.09 mmol) in THF (8 mL) and water (8 mL)
was cooled to 08C, and NaBH4 (0.69 g, 18.27 mmol) was added
portion-wise. The reaction mixture was stirred for 4 h, allowing the
reaction to warm to room temperature. The solvent was removed
under reduced pressure, water (20 mL) was added, and the mixture
was extracted with EtOAc. The combined organic layers were
washed with water and brine, dried over anhyd Na2SO4, filtered,
and concentrated. The crude product was purified on silica gel by
flash chromatography (hexanes/EtOAc, 100:0 to 80:20) to give 10
as a colorless oil (950 mg, 92%): UPLC–MS: tR =0.71 min, m/z:
149.0 [M+HꢁH2O].
3-(Bromomethyl)-6-isopropoxypyridazine (7c): Compound 7c
was prepared from 6c (489 mg, 3.21 mmol), according to the pro-
cedure described for the synthesis of 7a, and was obtained as an
orange oil (446 mg, 60%): 1H NMR (400 MHz, [D6]DMSO): d=7.53
(d, J=9.66 Hz, 1H), 6.96 (d, J=9.54 Hz, 1H), 5.12 (quin, J=6.63 Hz,
1H), 4.59 (s, 2H), 1.27 ppm (d, J=6.60 Hz, 6H); UPLC–MS: tR =
0.83 min, m/z: 232.9 [M+H].
4-Isopropoxybenzyl methyl carbonate (11): Methyl chloroformate
(0.920 mL, 11.91 mmol) was added slowly to a solution of 10
(900 mg, 5.41 mmol) in THF (30 mL) and pyridine (1.14 mL,
14.08 mmol) at 08C. The resulting white suspension was stirred
overnight at room temperature. A solution of aq HCl (6m) was
added until the mixture reached pH 1, and the mixture was then
extracted with methyl-tert-butyl ether. The combined organic
layers were washed with brine, dried over anhyd Na2SO4, filtered,
and concentrated. The crude product was purified on silica gel by
flash chromatography (hexanes/EtOAc, 100:0 to 90:10) to give 11
4-((6-Isopropoxypyridazin-3-yl)methyl)aniline (8c): Compound
8c was prepared from 7c (45 mg, 0.19 mmol), according to the
procedure described for the synthesis of 8a, and was obtained as
a yellow oil (34 mg, 56.7%): 1H NMR (400 MHz, [D6]DMSO): d=
7.60–7.67 (m, 2H), 7.52–7.58 (m, 1H), 7.19 (d, J=9.41 Hz, 1H), 6.89
(d, J=8.31 Hz, 2H), 6.80 (d, J=9.54 Hz, 1H), 6.50 (d, J=8.31 Hz,
2H), 5.12 (quin, J=6.63 Hz, 1H), 3.71 (s, 2H), 1.16–1.35 ppm (m,
6H); UPLC–MS: tR =0.72 min, m/z: 243.9 [M+H].
1
as a colorless oil (1.1 g, 90%): H NMR (400 MHz, CDCl3): d=7.18–
7.25 (m, 2H), 6.77–6.82 (m, 2H), 5.02 (s, 2H), 4.48 (dt, J=12.10,
6.05 Hz, 1H), 3.71 (s, 3H), 1.26 ppm (d, J=6.11 Hz, 6H); UPLC–MS:
tR =1.08 min.
N-(4-((6-Isopropoxypyridazin-3-yl)methyl)phenyl)-4,5-dihydro-
1H-imidazol-2-amine (9c): Compound 9c was prepared from 8c
(34 mg, 0.11 mmol), according to the procedure described for the
synthesis of 9a, and was obtained as a white powder (18 mg,
51.8%): 1H NMR (400 MHz, CD3OD): d=7.17 (d, J=9.41 Hz, 1H),
7.06 (d, J=8.31 Hz, 2H), 6.88 (d, J=8.31 Hz, 2H), 6.74 (d, J=
9.54 Hz, 1H), 5.14 (quin, J=6.63 Hz, 1H), 3.81 (s, 2H), 3.39 (s, 4H),
1.28 ppm (d, J=6.60 Hz, 6H); 13C NMR (151 MHz, CD3OD): d=
161.5, 161.1, 150.1, 147.8, 134.3, 133.0, 130.7 (2C), 130.1, 124.6 (2C),
50.8, 44.0 (2C), 41.3, 21.3 ppm (2C); UPLC–MS: tR =0.56 min, m/z:
311.8 [M+H].
2-Fluoro-4-(4-isopropoxybenzyl)aniline (12a): A solution of 11
(200 mg, 0.89 mmol) and 3-fluoro-4-(4,4,5,5-tetramethyl-1,3,2-dioxa-
borolan-2-yl)aniline (233 mg, 0.981 mmol) in DMF (2 mL) was de-
gassed using an argon stream for 10 min. Potassium carbonate
(370 mg, 2.68 mmol), allylpalladium chloride dimer (48.9 mg,
0.13 mmol), and 1,5-bis(diphenylphosphino)pentane (118 mg,
0.27 mmol) were added to the solution at room temperature
&
ChemMedChem 2016, 11, 1 – 14
www.chemmedchem.org
8
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
under argon, then the suspension was stirred at 658C for 16 h.
After cooling to room temperature, EtOAc (20 mL) and water
(20 mL) were added, and the biphasic suspension was filtered over
Celite. The filter cake was washed with EtOAc and water. The or-
ganic layer was washed with brine, dried over anhyd Na2SO4, fil-
tered, and concentrated. The crude product was purified on silica
gel by flash chromatography (hexanes/EtOAc, 100:0 to 90:10) to
provide 12a as a yellow oil (130 mg, 51.7%): 1H NMR (400 MHz,
CDCl3): d=7.08 (d, J=8.56 Hz, 2H), 6.69–6.87 (m, 5H), 4.52 (quin,
J=6.05 Hz, 1H), 3.82 (s, 2H), 1.34 ppm (d, J=5.99 Hz, 6H); UPLC–
MS: tR =1.19 min, m/z: 260.1 [M+H].
4.56 (dt, J=12.07, 6.01 Hz, 1H), 4.03 (s, 2H), 3.52 (s, 4H), 1.30 ppm
(d, J=5.99 Hz, 6H); UPLC–MS: tR =1.01 min, m/z: 378.5 [M+H].
4-(4-Isopropoxybenzyl)-2-methoxyaniline (12d): Compound 12d
was prepared from 11 (100 mg, 0.45 mmol) and 2-methoxy-4-
(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)aniline,[21] according to
the procedure described for the synthesis of 12a, and was ob-
1
tained as a yellow oil (56 mg, 38%): H NMR (400 MHz, [D6]DMSO):
d=7.08 (d, J=8.56 Hz, 2H), 6.80 (d, J=8.56 Hz, 2H), 6.66 (d, J=
1.10 Hz, 1H), 6.40–6.58 (m, 2H), 4.51–4.57 (m, 1H), 3.72 (s, 3H),
3.70 (s, 2H), 1.23 ppm (d, J=5.99 Hz, 6H); UPLC–MS: tR =1.17 min,
m/z: 272.4 [M+H].
N-(2-Fluoro-4-(4-isopropoxybenzyl)phenyl)-4,5-dihydro-1H-imi-
dazol-2-amine (13a): A suspension of 12a (100 mg, 0.39 mmol)
and 2-chloro-4,5-dihydro-1H-imidazole (69.3 mg, 0.66 mmol) in 2-
propanol (4 mL) was stirred at reflux for 3 h under argon. The reac-
tion mixture was concentrated, and the residue was extracted with
methyl-tert-butyl ether and aq NaOH (2m), the organic layer was
washed with brine, dried over anhyd Na2SO4, filtered, and concen-
trated. The crude product was purified on silica gel (CH2Cl2/MeOH/
NH4OH, 99:1:1 to 80:20:1) to provide 13a as a white powder
(45 mg, 35%): 1H NMR (400 MHz, CDCl3): d=7.10 (d, J=8.56 Hz,
2H), 6.96 (t, J=8.25 Hz, 1H), 6.80–6.89 (m, 4H), 4.53 (dt, J=12.13,
6.10 Hz, 2H), 3.85 (s, 2H), 3.55 (s, 4H), 1.53–1.77 (brs, 2H),
1.35 ppm (d, J=5.99 Hz, 6H); UPLC–MS: tR =0.83 min, m/z: 328.4
[M+H].
N-(4-(4-Isopropoxybenzyl)-2-methoxyphenyl)-4,5-dihydro-1H-
imidazol-2-amine (13d): Compound 13d was prepared from 12d
(55 mg, 0.17 mmol), according to the procedure described for the
synthesis of 13a, and obtained as a yellow oil (23 mg, 40.4%):
1H NMR (400 MHz, CD3OD): d=7.11 (d, J=8.68 Hz, 2H), 7.01 (d, J=
7.95 Hz, 1H), 6.78–6.85 (m, 3H), 6.74 (dd, J=7.95, 1.59 Hz, 1H),
4.55 (dt, J=12.10, 6.05 Hz, 1H), 3.86 (s, 2H), 3.78 (s, 3H), 3.50 (s,
4H), 1.30 ppm (d, J=6.11 Hz, 6H); UPLC–MS: tR =0.90 min, m/z:
340.4 [M+H].
4-Isopropoxypyridin-2-ol
(14):
2-Bromopropane
(0.84 mL,
9.0 mmol) and K2CO3 (1.49 g, 10.8 mmol) were added to a solution
of pyridine-2,4-diol (1.0 g, 9.0 mmol) in anhydrous DMF (10 mL) at
room temperature under argon. The reaction mixture was stirred
overnight at 658C. The off-white suspension was cooled to room
temperature, water was added, and the product extracted with
CH2Cl2/2-propanol (70:30). The organic layer was dried over
a phase separator and concentrated. The crude product was puri-
fied on silica gel by flash chromatography (hexanes/EtOAc, 50:50
to 0:100 then CH2Cl2/MeOH 100:0 to 80:20) to provide 14 as
a white powder (668 mg, 48%), in addition to 2-isopropoxypyridin-
4-ol (175 mg, 13%): 1H NMR (400 MHz, [D6]DMSO): d=11.02 (brs,
1H), 7.20 (d, J=7.34 Hz, 1H), 5.79 (dd, J=7.34, 2.32 Hz, 1H), 5.65
(d, J=2.32 Hz, 1H), 4.58 (dt, J=12.01, 6.04 Hz, 1H), 1.24 ppm (d,
J=5.99 Hz, 6H); UPLC–MS: tR =0.57 min, m/z: 154.0 [M+H].
3-Fluoro-4-(4-isopropoxybenzyl)aniline (12b): Compound 12b
was prepared from 11 (200 mg, 0.89 mmol), according to the pro-
cedure described for the synthesis of 12a. After purification by
flash chromatography, 12b was obtained as a yellow oil (130 mg,
51.7%): 1H NMR (400 MHz, CDCl3): d=7.08 (m, J=8.44 Hz, 2H),
6.89 (t, J=8.19 Hz, 1H), 6.77–6.82 (m, J=8.44 Hz, 2H), 6.38 (d, J=
9.66 Hz, 2H), 4.49 (dt, J=12.10, 6.05 Hz, 1H), 3.81 (s, 2H), 3.65 (brs,
2H), 1.31 ppm (d, J=6.11 Hz, 6H); UPLC–MS: tR =1.17 min, m/z:
260.1 [M+H].
N-(3-Fluoro-4-(4-isopropoxybenzyl)phenyl)-4,5-dihydro-1H-imi-
dazol-2-amine (13b): Compound 13b was prepared from 12b
(120 mg, 0.46 mmol), according to the procedure described for the
synthesis of 13a. After purification by flash chromatography, 13b
was obtained as a yellow oil (105 mg, 68.8%): 1H NMR (400 MHz,
CDCl3): d=7.09–7.14 (m, 2H), 6.99 (t, J=8.13 Hz, 1H), 6.78–6.83 (m,
J=8.44 Hz, 2H), 6.64–6.70 (m, 2H), 4.46–4.55 (m, 2H), 3.86 (s, 2H),
3.52 (s, 4H), 1.64 (brs, 2H), 1.32 (d, J=6.11 Hz, 6H); 13C NMR
(151 MHz, CDCl3): d=161.3 (d, J=238 Hz, 1C), 157.9, 156.2, 150.1,
132.4, 131.1, 129.7 (2C), 121.6, 118.7, 115.8 (2C), 109.8, 69.9, 42.4,
33.6 (2C), 22.2 ppm (2C); UPLC–MS: tR =0.84 min, m/z: 328.4 [M+
H].
4-Isopropoxy-1-(4-nitrobenzyl)pyridin-2(1H)-one (15a): K2CO3
(361 mg,
2.61 mmol)
and
1-(bromomethyl)-4-nitrobenzene
(367 mg, 1.697 mmol) were slowly added to a solution of 14
(200 mg, 1.306 mmol) in THF (8 mL). The reaction mixture was
stirred overnight at reflux. The mixture was cooled to room tem-
perature, water was added, and the product was extracted with
methyl-tert-butyl ether. The organic layer was dried over a phase
separator and concentrated to provide 15a as a white powder
1
(285 mg, 75%): H NMR (400 MHz, CDCl3): d=8.19 (m, J=8.56 Hz,
2H), 7.43 (d, J=8.56 Hz, 2H), 7.13 (d, J=7.21 Hz, 1H), 5.89–5.94 (m,
2H), 5.16 (s, 2H), 4.52 (dt, J=12.17, 6.02 Hz, 1H), 1.35 ppm (d, J=
6.11 Hz, 6H); UPLC–MS: tR =0.93 min, m/z: 289.3 [M+H].
4-(4-Isopropoxybenzyl)-3-(trifluoromethyl)aniline (12c): Com-
pound 12c was prepared from 11 (100 mg, 0.45 mmol), according
to the procedure described for the synthesis of 12a. After purifica-
tion by flash chromatography, 12c was obtained as a yellow oil
(65 mg, 43.8%): 1H NMR (400 MHz, [D6]DMSO): d=6.92–7.02 (m,
3H), 6.88 (d, J=2.32 Hz, 1H), 6.81 (d, J=8.56 Hz, 2H), 6.73 (dd, J=
8.31, 2.08 Hz, 1H), 5.40 (s, 2H), 4.53 (dt, J=12.07, 6.01 Hz, 1H), 3.85
(s, 2H), 1.24 ppm (d, J=6.11 Hz, 6H); UPLC–MS: tR =1.30 min, m/z:
310.0 [M+H].
1-(4-Aminobenzyl)-4-isopropoxypyridin-2(1H)-one (16a): A solu-
tion of 15a (71 mg, 0.25 mmol), triethylamine (0.34 mL, 2.46 mmol)
and Pd/C (14.7 mg, 0.014 mmol) in MeOH (5.0 mL) was stirred at
room temperature under hydrogen (0.1 bar) for 2 h. The reaction
mixture was filtered over Celite, and the Celite pad was washed
with EtOAc. The filtrate was concentrated, and the residue was dis-
solved with EtOAc and washed with water, dried over a phase sep-
arator, and concentrated. The crude product was purified on silica
gel by flash chromatography (cyclohexane/EtOAc, 100:0 to 0:100)
to afford 16a as a colorless oil (55 mg, 86%): 1H NMR (400 MHz,
[D6]DMSO): d=7.54 (d, J=7.58 Hz, 1H), 6.99 (d, J=8.31 Hz, 2H),
6.49 (d, J=8.31 Hz, 2H), 5.86 (dd, J=7.58, 2.57 Hz, 1H), 5.76 (d, J=
2.57 Hz, 1H), 5.05 (s, 2H), 4.80 (s, 2H), 4.58 (dt, J=11.98, 5.99 Hz,
N-(4-(4-Isopropoxybenzyl)-3-(trifluoromethyl)phenyl)-4,5-dihy-
dro-1H-imidazol-2-amine (13c): Compound 13c was prepared
from 12c (65 mg, 0.19 mmol), according to the procedure de-
scribed for the synthesis of 13a, and was obtained as a yellow oil
1
(65 mg, 86%): H NMR (400 MHz, CD3OD): d=7.30 (s, 1H), 7.14 (d,
J=0.98 Hz, 2H), 7.05 (d, J=8.56 Hz, 2H), 6.83 (d, J=8.68 Hz, 2H),
ChemMedChem 2016, 11, 1 – 14
www.chemmedchem.org
9
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
1H), 1.24 ppm (d, J=5.99 Hz, 6H); UPLC–MS: tR =0.73 min, m/z:
259.1 [M+H].
0.84 ppm (d, J=6.72 Hz, 6H); UPLC–MS: tR =0.87 min, m/z: 232.0
[M+2].
1-(4-((4,5-Dihydro-1H-imidazol-2-yl)amino)benzyl)-4-isopropoxy-
pyridin-2(1H)-one (17a): A mixture of 16a (55 mg, 0.21 mmol) and
2-chloro-4,5-dihydro-1H-imidazole (59.9 mg, 0.28 mmol) in isopro-
panol (2.8 mL) was stirred at 858C in a sealed glass tube for 1 h
under an argon atmosphere. The reaction mixture was cooled to
room temperature, concentrated, and the residue extracted with
CH2Cl2/aq NaOH (2m), the organic layer was washed with brine,
dried over a phase separator, and concentrated. The crude product
was purified on silica gel by flash chromatography (CH2Cl2/MeOH/
14.7m NH4OH, 100:0:0 to 90:10:1) to afford 17a as a white foam
(53 mg, 76%): 1H NMR (400 MHz, CD3OD): d=7.55 (d, J=7.58 Hz,
1H), 7.24 (d, J=8.31 Hz, 2H), 7.00 (d, J=8.31 Hz, 2H), 6.06 (dd, J=
7.58, 2.69 Hz, 1H), 5.93 (d, J=2.69 Hz, 1H), 5.06 (s, 2H), 4.53–4.71
(m, 1H), 3.51 (s, 4H), 1.34 ppm (d, J=5.99 Hz, 6H); 1C NMR
(400 MHz, CD3OD): d=168.8, 166.4, 161.2, 149.0, 139.9, 132.0, 130.0
(2C), 124.6 (2C), 103.6, 99.0, 72.1, 52.3, 43.9 (2C), 21.9 ppm (2C);
UPLC–MS: tR =0.62 min, m/z: 327.2 [M+H].
1-Isobutyl-4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)pyridin-
2(1H)-one (20): A mixture of 19 (224 mg, 0.97 mmol), bis(pinacola-
to)diboron (49 mg, 1.95 mmol), and KOAc (287 mg, 2.92 mmol) in
1,4-dioxane (5.0 mL) was degassed with argon for 5 min.
PdCl2(dppf)/CH2Cl2 adduct (79 mg, 0.09 mmol) was added, and the
vial was capped. The reaction mixture was heated in a microwave
at 1208C for 1 h. The reaction mixture was filtered over Celite and
concentrated. The crude product was purified by flash chromatog-
raphy (cyclohexane/EtOAc, 100:0 to 0:100) to give 20 as brown oil
(320 mg, 95%): 1H NMR (400 MHz, [D6]DMSO): d=7.58 (d, J=
6.72 Hz, 1H), 6.61 (s, 1H), 6.26 (d, J=6.72 Hz, 1H), 3.69 (d, J=
7.34 Hz, 2H), 1.94–2.13 (m, 1H), 1.25–1.30 (m, 12H), 0.84 ppm (d,
J=6.72 Hz, 6H); UPLC–MS: tR =0.54 min.
1-Isobutyl-4-(4-nitrobenzyl)pyridin-2(1H)-one (21): PdCl2(dppf)/
CH2Cl2 adduct (56.7 mg, 0.069 mmol) was added to a solution of
20
(289 mg,
0.833 mmol),
1-(bromomethyl)-4-nitrobenzene
(150 mg, 0.694 mmol), and K2CO3 (221 mg, 2.083 mmol) in dime-
thoxyethane (2.3 mL), and EtOH/water (0.46 mL, 1:1) was added at
room temperature under argon. Argon was bubbled through the
stirred reaction mixture for 5 min before microwaving at 1208C
while stirring for 10 min. The reaction mixture was filtered over
Celite, the filtrate was concentrated, and the crude product was
purified on silica gel by flash chromatography (cyclohexane/EtOAc,
100:0 to 30:70) to afford 21 as an orange oil (67 mg, 33.7%):
1H NMR (400 MHz, [D6]DMSO): d=8.19 (d, J=8.44 Hz, 2H), 7.55
(dd, J=10.70, 7.76 Hz, 3H), 6.25 (s, 1H), 6.08 (d, J=6.97 Hz, 1H),
3.92 (s, 2H), 3.60–3.68 (m, 2H), 2.01 (dt, J=13.79, 6.86 Hz, 1H),
0.82 ppm (d, J=6.72 Hz, 6H); UPLC–MS: tR =0.96 min, m/z: 287.1
[M+H].
1-(2-Fluoro-4-nitrobenzyl)-4-isopropoxypyridin-2(1H)-one (15b):
Compound 15b was prepared from 14 (50 mg, 0.33 mmol) and 1-
(bromomethyl)-2-fluoro-4-nitrobenzene,[22] according to the proce-
dure described for the synthesis of 15a (86 mg, 85%): 1H NMR
(400 MHz, CDCl3): d=8.00 (d, J=8.75 Hz, 1H), 7.94 (d, J=9.69 Hz,
1H), 7.59 (t, J=7.89 Hz, 1H), 7.19–7.25 (m, 1H), 5.86–5.93 (m, 2H),
5.15 (s, 2H), 4.50 (dt, J=12.17, 6.02 Hz, 1H), 1.34 ppm (d, J=
5.99 Hz, 6H); UPLC–MS: tR =0.95 min, m/z: 307.4 [M+H].
1-(4-Amino-2-fluorobenzyl)-4-isopropoxypyridin-2(1H)-one
(16b): A saturated solution of Cu(OAc)2 (47,4 mg, 0.26 mmol) in
MeOH (2.5 mL) was added to a solution of 15b (80 mg, 0.26 mmol)
in MeOH (7 mL). The reaction mixture was cooled to 08C and
NaBH4 (148 mg, 3.92 mmol) was added in four portions. The reac-
tion mixture was stirred at 08C for 15 min and filtered over Celite,
then the filtrate was concentrated. The residue was dissolved in
EtOAc, washed with water, dried over a phase separator, and con-
centrated. The crude product was purified on silica gel by flash
chromatography (hexanes/EtOAc, 50:50 to 0:100) to provide 16b
4-(4-Aminobenzyl)-1-isobutylpyridin-2(1H)-one (22): A solution of
21 (67 mg, 0.23 mmol), triethylamine (0.33 mL, 2.34 mmol), and Pd/
C (12.45 mg, 0.012 mmol) in EtOH (5.00 mL) was stirred at room
temperature under H2 atmosphere (0.1 bar) for 2 h. The reaction
mixture was filtered over Celite, and the Celite pad was washed
with EtOH. The filtrate was concentrated, and the crude product
was purified on silica gel by flash chromatography (CH2Cl2/MeOH/
14.7m NH4OH, 99:1:0.1 to 90:10:1) to afford 22 as a yellow oil
(54 mg, 90%): 1H NMR (600 MHz, [D6]DMSO): d=7.47 (d, J=
6.97 Hz, 1H), 6.88 (d, J=8.25 Hz, 2H), 6.50 (d, J=8.44 Hz, 2H), 6.12
(s, 1H), 6.01 (dd, J=6.97, 1.83 Hz, 1H), 4.94 (s, 2H), 3.62 (d, J=
7.34 Hz, 2H), 3.54 (s, 2H), 1.94–2.05 (m, 1H), 0.82 ppm (d, J=
6.79 Hz, 6H); UPLC–MS: tR =0.73 min, m/z: 257.1 [M+H].
1
as colorless oil (34 mg, 46.6%): H NMR (400 MHz, CDCl3): d=7.17–
7.26 (m, 2H), 6.34–6.43 (m, 2H), 5.79–5.88 (m, 2H), 4.98 (s, 2H),
4.48 (dt, J=12.13, 6.10 Hz, 1H), 3.79 (brs, 2H), 1.32 ppm (d, J=
5.99 Hz, 6H); UPLC–MS: tR =0.82 min, m/z: 277.1 [M+H].
1-(4-((4,5-Dihydro-1H-imidazol-2-yl)amino)-2-fluorobenzyl)-4-iso-
propoxypyridin-2(1H)-one (17b): Compound 17b was prepared
from 16b (30 mg, 0.11 mmol), according to the procedure de-
1
scribed for the synthesis of 17a (29 mg, 77%): H NMR (400 MHz,
4-(4-((4,5-Dihydro-1H-imidazol-2-yl)amino)benzyl)-1-isobutylpyri-
din-2(1H)-one (23): A mixture of 22 (54 mg, 0.21 mmol) and 2-
chloro-4,5-dihydro-1H-imidazole (59.3 mg, 0.28 mmol) in isopropa-
nol (2.8 mL) was stirred at 858C in a sealed glass tube for 40 min
under argon. After cooling to room temperature, the reaction mix-
ture was evaporated, and the residue was extracted with CH2Cl2
and aq NaOH (2m). The combined organic layers were washed
with brine, dried over a phase separator, and concentrated. The
crude product was purified on silica gel by flash chromatography
(CH2Cl2/MeOH/14.7m NH4OH, 100:0:0 to 90:10:1) to afford 23 as
CDCl3): d=7.12–7.26 (m, 2H), 6.59–6.68 (m, 2H), 5.74–5.81 (m, 2H),
4.95 (s, 2H), 4.36–4.45 (m, 1H), 3.46 (s, 4H), 1.25 ppm (d, J=
6.11 Hz, 6H); UPLC–MS: tR =0.59 min, m/z: 345.4 [M+H].
4-Bromo-1-isobutylpyridin-2(1H)-one (19): A suspension of 4-bro-
mopyridin-2(1H)-one (0.5 g, 2.87 mmol), 1-bromo-2-methylpropane
(0.37 mL, 3.45 mmol), K2CO3 (0.67 g, 4.89 mmol), and NaI (0.02 g,
0.14 mmol) in DMF (5.01 mL) was stirred in a sealed glass tube at
608C for 16 h under argon. The reaction mixture was diluted with
EtOAc at room temperature. The organic layer was washed with
water and brine, dried over a phase separator, and concentrated.
The crude product was purified on silica gel by flash chromatogra-
phy (cyclohexane/EtOAc, 100:0 to 0:100) to afford the major prod-
uct 19 as a colorless oil (407 mg, 61%), and 4-bromo-2-isobutoxy-
1
a white foam (27 mg, 38.7%): H NMR (400 MHz, CD3OD): d=7.48
(d, J=6.97 Hz, 1H), 7.17 (d, J=8.07 Hz, 2H), 7.01 (d, J=8.19 Hz,
2H), 6.36 (s, 1H), 6.27 (d, J=6.85 Hz, 1H), 3.70–3.86 (m, 4H), 3.48–
3.57 (m, 4H), 2.13 (dt, J=13.63, 6.88 Hz, 1H), 0.93 ppm (d, J=
6.72 Hz, 6H); UPLC–MS:]0.62 min, m/z: 325.1 [M+H].
1
pyridine (120 mg, 18%): H NMR (400 MHz, [D6]DMSO): d=7.64 (d,
J=7.21 Hz, 1H), 6.71 (d, J=2.20 Hz, 1H), 6.45 (dd, J=7.21, 2.20 Hz,
1H), 3.69 (d, J=7.46 Hz, 2H), 2.03 (dt, J=13.60, 6.59 Hz, 1H),
1-(Bromomethyl)-4-isopropoxybenzene (24): Polymer-bound tri-
phenylphosphine, (2.4 g, 7.67 mmol) and CBr4 (2.54 g, 7.67 mmol)
&
ChemMedChem 2016, 11, 1 – 14
www.chemmedchem.org
10
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
were added to a solution of 10 (1.0 g, 5.11 mmol) in CH2Cl2 (17 mL)
at 08C under argon. The reaction mixture was allowed to warm to
room temperature and was stirred for 1 h, then filtered over Celite,
and the Celite pad was washed with CH2Cl2. The filtrate was con-
centrated, and the crude product was purified on silica gel by flash
chromatography (cyclohexane/EtOAc, 100:0 to 95:5) to afford 24
as an orange oil (842 mg, purity 60%), which was used in the next
step without further purification.
water and dried to give 135 mg of 28a as a yellow solid, which
was used in the next step without further purification.
N-(4-(4-Isopropoxybenzyl)phenyl)-4,5-dihydrothiazol-2-amine
(29a): Compound 28a (135 mg, 0.449 mmol) in EtOH (1.8 mL) was
added to a solution of 2-bromoethanamine hydrobromide (92 mg,
0.45 mmol) in water (0.22 mL) at room temperature under argon.
The reaction mixture was stirred at reflux for 16 h. After cooling to
room temperature, the EtOH was evaporated, and the residue was
basified with diluted aq ammonia and extracted with CH2Cl2. The
combined organic layers were dried over a phase separator and
concentrated. The crude product was purified on silica gel by flash
chromatography (cyclohexane/EtOAc, 100:0 to 80:20) to afford
29a (34 mg, 22.9%) as a white solid: 1H NMR (400 MHz, CD3OD):
d=7.01–7.18 (m, 6H), 6.77–6.86 (m, 2H), 4.55 (dt, J=12.10,
6.05 Hz, 1H), 3.82–3.91 (m, 4H), 3.28–3.31 (m, 2H), 1.30 ppm (d, J=
6.11 Hz, 6H); 13C NMR (151 MHz, CD3OD): d=157.6, 147.4, 138.0,
135.1, 130.8 (2C), 130.2 (2C), 122.9 (2C), 117.1 (2C), 121.9, 71.0, 41.4,
33.0, 30.8, 22.4 ppm (2C); UPLC–MS: tR =0.94 min, m/z: 327.4 [M+
H].
4-Chloro-1-(4-isopropoxybenzyl)pyridin-2(1H)-one (25): A mixture
of 4-chloropyridin-2(1H)-one (50 mg, 0.39 mmol), 24 (115 mg,
0.50 mmol), and K2CO3 (107 mg, 0.77 mmol) in THF (2.5 mL) was
heated in a microwave at 1308C for 30 min. After cooling to room
temperature, water was added to the reaction mixture, and the
product was extracted with CH2Cl2. The organic layer was dried
over a phase separator and concentrated, and the crude product
was purified on silica gel by flash chromatography (cyclohexane/
EtOAc, 100:0 to 70:30) to afford 25 as a colorless oil (60 mg, 56%):
1H NMR (400 MHz, CDCl3): d=7.13 (dd, J=12.90, 7.89 Hz, 3H), 6.78
(d, J=8.56 Hz, 2H), 6.53–6.60 (m, 1H), 6.08 (dd, J=7.27, 2.26 Hz,
1H), 4.95 (s, 2H), 4.46 (dt, J=12.07, 6.01 Hz, 1H), 1.25 ppm (d, J=
6.11 Hz, 6H); UPLC–MS: tR =1.05 min, m/z: 278.1 [M+H].
1-(3-Fluoro-4-(4-isopropoxybenzyl)phenyl)thiourea (28b): Com-
pound 28b was prepared from 12b (200 mg, 0.77 mmol), accord-
1
4-((4,5-Dihydro-1H-imidazol-2-yl)amino)-1-(4-isopropoxybenzyl)-
pyridin-2(1H)-one (26): 4,5-Dihydro-1H-imidazol-2-amine (34.6 mg,
0.234 mmol) and LHMDS (1m in THF) (0.42 mL, 0.42 mmol) were
ing to the procedure described for the synthesis of 28a: H NMR
(400 MHz, CDCl3): d=8.07 (brs, 1H), 7.20 (t, J=8.09 Hz, 1H), 7.08–
7.13 (m, 2H), 6.90–7.00 (m, 2H), 6.80–6.85 (m, 2H), 6.14 (brs, 2H),
4.46–4.56 (m, 1H), 3.92 (s, 2H), 1.32 ppm (d, J=5.99 Hz, 6H);
UPLC–MS: tR =1.06 min, m/z: 319.1 [M+H].
added to
a
solution of [BrettPhos-PdII] precatalyst (8.48 mg,
10.62 mmol), BrettPhos (5.70 mg, 10.62 mmol), and 25 (59 mg,
0.212 mmol) in anhydrous THF (2.1 mL) at room temperature
under argon. The reaction mixture was heated in a microwave at
1308C for 30 min, then cooled to 08C, quenched with saturated aq
NH4Cl, diluted with water, and extracted with EtOAc. The organic
layer was dried over a phase separator and concentrated, and the
crude product was purified on silica gel by flash chromatography
(CH2Cl2/MeOH/14.7m NH4OH, 100:0:0 to 50:50:5). The fractions
were concentrated, and the residue was treated with 5% MeOH in
CH2Cl2, filtered, and concentrated to give 26 as a colorless oil
N-(3-Fluoro-4-(4-isopropoxybenzyl)phenyl)-4,5-dihydrothiazol-2-
amine (29b): Compound 29b was prepared from 28b (100 mg,
0.31 mmol), according to the procedure described for the synthesis
of 29a: 1H NMR (400 MHz, CDCl3): d=7.03 (d, J=8.31 Hz, 2H),
6.82–6.97 (m, 2H), 6.67–6.76 (m, 3H), 4.42 (dt, J=12.07, 6.01 Hz,
1H), 3.80 (s, 2H), 3.71–3.77 (m, 2H), 3.25 (t, J=7.03 Hz, 2H),
1.22 ppm (d, J=6 Hz, 6H); UPLC–MS: tR =0.96 min, m/z: 345.4 [M+
H].
1
(5 mg, 7%): H NMR (400 MHz, CD3OD): d=7.63 (d, J=7.70 Hz, 1H),
1-(2-Chloroethyl)-3-(4-(4-isopropoxybenzyl)phenyl)urea (30): 1-
Chloro-2-isocyanatoethane (0.13 mL, 1.49 mmol) was added drop-
wise to a solution of 27a (300 mg, 1.24 mmol) in CH2Cl2 (5 mL) at
room temperature under argon. The reaction mixture was stirred
overnight at room temperature, and the precipitate was collected
by filtration. The filter cake was washed with cold Et2O and dried.
The product was obtained as a white powder (305 mg, 70%) and
used in the next step without further purification: UPLC–MS: tR =
1.16 min, m/z: 347.1 [M+H].
7.27 (d, J=8.56 Hz, 2H), 6.83–6.97 (m, 3H), 6.25 (d, J=2.57 Hz, 1H),
5.08 (s, 2H), 4.59 (quin, J=5.99 Hz, 1H), 3.85–3.99 (m, 2H), 3.58 (t,
J=8.01 Hz, 2H), 1.30 ppm (d, J=5.99 Hz, 6H); UPLC–MS: tR =
0.65 min, m/z: 327.2 [M+H].
1-(4-(4-Isopropoxybenzyl)phenyl)thiourea (28a): A solution of
benzoyl isothiocyanate (0.11 mL, 0.83 mmol) in anhydrous acetone
(1.6 mL) was added dropwise to a stirred solution of 27a (200 mg,
0.83 mmol) in anhydrous acetone (2.5 mL) at room temperature
under argon. The reaction mixture was stirred at room temperature
for 1 h and concentrated. The residue was filtered and washed
with a small amount of cold anhydrous acetone, and the residual
solvent was evaporated. The crude product was purified on silica
gel by flash chromatography (cyclohexane/EtOAc, 100:0 to 90:10)
to afford N-((4-(4-isopropoxybenzyl)phenyl)carbamothioyl)benza-
mide (258 mg, 77%) as a white solid: 1H NMR (400 MHz, CDCl3):
d=12.45 (brs, 1H), 8.97 (brs, 1H), 7.82 (d, J=7.58 Hz, 2H), 7.53–
7.65 (m, 3H), 7.44–7.52 (m, 2H), 7.16 (d, J=8.31 Hz, 2H), 7.02 (d,
J=8.44 Hz, 2H), 6.75 (d, J=8.56 Hz, 2H), 4.44 (dt, J=12.10,
6.05 Hz, 1H), 3.86 (s, 2H), 1.25 ppm (d, J=5.99 Hz, 6H); UPLC–MS:
tR =1.45 min, m/z: 405.1 [M+H]. This intermediate (250 mg,
0.62 mmol) was then added in one portion to a 5% aq NaOH
(1 mL) solution while stirring at 908C. The reaction mixture was
stirred at 908C for 20 min, cooled to room temperature, and acidi-
fied with 15% aq HCl. The pH was adjusted to 8.0 using aq ammo-
nia, and the crude product was filtered. The solid was washed with
N-(4-(4-Isopropoxybenzyl)phenyl)-4,5-dihydrooxazol-2-amine
(31): A mixture of SiO2·KF (60:40) (511 mg, 4.32 mmol) was added
to a solution of 30 (150 mg, 0.43 mmol) in CH3CN (6 mL) at room
temperature under argon. The reaction mixture was stirred at
reflux for 4 h, cooled to room temperature and filtered, and the
solvent was evaporated. The crude product was purified by prepa-
rative HPLC (column: Waters Sunfire C18, 5 mm, 30ꢁ100 mm, sol-
vent A: H2O+0.1%TFA, solvent B: CH3CN+0.1%TFA, flow:
50 mLminꢁ1, gradient: 20 to 40% B over 16 min) to afford 31 as
a white powder (130 mg, 69.4%): 1H NMR (400 MHz, [D6]DMSO):
d=7.21–7.34 (m, 4H), 7.13 (d, J=8.31 Hz, 2H), 6.83 (d, J=8.44 Hz,
2H), 4.85 (t, J=8.56 Hz, 2H), 3.88 (s, 2H), 3.81–3.86 (m, 2H),
1.24 ppm (d, J=5.99 Hz, 6H); UPLC–MS: tR =0.88 min, m/z: 311.4
[M+H].
N-(3-Fluoro-4-(4-isopropoxybenzyl)phenyl)-3,4-dihydro-2H-
pyrrol-5-amine (32a):
A solution of phosphorus oxychloride
(0.3 mL, 3.20 mmol) in toluene (2 mL) was added dropwise to a so-
ChemMedChem 2016, 11, 1 – 14
www.chemmedchem.org
11
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
lution of pyrrolidin-2-one (0.31 mL, 4.0 mmol) in toluene (2 mL) at
08C under argon, and the reaction mixture was stirred overnight at
room temperature. A solution of 12b (50 mg, 0.400 mmol) in tolu-
ene (2 mL) was added, and the reaction mixture was stirred at
reflux for 2 h. MeOH (0.5 mL) was added at room temperature. The
product was extracted with 10% aq NaOH and EtOAc, and the
combined organic layers were dried over a phase separator and
concentrated. The crude product was purified on silica gel by flash
chromatography (CH2Cl2/MeOH/14.7m NH4OH, 99:1:0.1 to 90:10:1)
to give 32a as colorless oil (35 mg, 26.6%): 1H NMR (400 MHz,
CDCl3): d=7.04 (d, J=8.31 Hz, 2H), 6.94 (t, J=8.38 Hz, 1H), 6.58–
6.76 (m, 4H), 4.42 (dt, J=12.10, 6.05 Hz, 1H), 3.79 (s, 2H), 3.32–3.46
(m, 2H), 2.48 (t, J=7.89 Hz, 2H), 1.94–2.05 (m, 2H), 1.24 ppm (d,
J=5.99 Hz, 6H); UPLC–MS: tR =0.88 min, m/z: 327.2 [M+H].
at 378C in assay binding buffer (final volume 0.5 mL) with gentle
agitation for 2 h to ensure equilibrium was reached. [3H]NMS bind-
ing to CHO-M3 membranes (10 mg per well) was similarly carried
out in HBSS buffer at 378C (final volume 1.5 mL) with gentle agita-
tion for 2 h. Following this incubation period, bound and free
radioligand were separated by rapid vacuum filtration using a Filter-
Mate Cell Harvester (PerkinElmer, Beaconsfield, UK) onto 96-well
GF/B filter plates pre-treated with assay buffer and rapidly washed
three times with ice-cold 20 mm HEPES containing 1 mm MgCl2,
pH 7.4. After drying (>4 h), 40 mL of Microscint 20 (PerkinElmer)
was added to each well, and radioactivity was quantified using
single photon counting on a TopCount microplate scintillation
counter (PerkinElmer). Aliquots of [3H]1 and [3H]NMS were also
quantified accurately to determine how much radioactivity was
added to each well using liquid scintillation spectrometry on
a Hidex 300SL scintillation counter (LabLogic, Sheffield, UK). In all
experiments, total binding never exceeded more than 10% of
ligand added, therefore limiting significant depletion of the free
radioligand concentration.[24]
N-(3-Fluoro-4-(4-(3-fluoropropoxy)benzyl)phenyl)-3,4-dihydro-
2H-pyrrol-5-amine (32b): Compound 32b was prepared from 4b
(70 mg, 0.25 mmol) according to the procedure described for the
synthesis of 32a (50 mg, 55.8%): 1H NMR (400 MHz, CDCl3): d=
7.11–7.16 (m, 2H), 7.01 (brt, J=8.19 Hz, 1H), 6.84 (brd, J=7.46 Hz,
2H), 6.75 (brd, J=9.41 Hz, 1H), 6.67 (brd, J=8.07 Hz, 1H), 4.64 (dt,
J=48 Hz, 5.32 Hz, 2H), 4.08 (brt, J=5.69 Hz, 2H), 3.88 (s, 2H),
3.41–3.50 (m, 2H), 2.42–2.59 (m, 2H), 2.02–2.22 ppm (m, 3H, 1.25
(brt, J=6.79 Hz, 1H); UPLC–MS: tR =0.83 min, m/z: 345.2 [M+H].
Data analysis: [3H]1 saturation binding assays were analyzed by
nonlinear regression using Prism 6.0 (GraphPad Software, San
Diego, USA). Specific binding, obtained by subtracting NSB from
total binding, was fitted to a one-site binding model which de-
scribes a rectangular hyperbola or binding isotherm, leading to in-
dividual estimates for total receptor number and radioligand equi-
librium dissociation constant.
N-(3-Fluoro-4-(4-isopropoxybenzyl)phenyl)-3,4,5,6-tetrahydro-
pyridin-2-amine (33a): A solution of phosphorus oxychloride
(0.14 mL, 1.54 mmol) in toluene (2 mL) was added dropwise to a so-
lution of piperidin-2-one (191 mg, 1.93 mmol) in toluene (2 mL) at
08C under argon. The reaction mixture was stirred at room temper-
ature overnight. A solution of 12b (50 mg, 0.193 mmol) in toluene
(2 mL) was added, and the reaction mixture was stirred at reflux
for 2 h. After adding MeOH (0.2 mL) at room temperature, the
product was extracted with 10% aq NaOH and methyl-tert-butyl
ether. The combined organic layers were dried over a phase sepa-
rator and concentrated. The crude product was purified on silica
gel by flash chromatography (CH2Cl2/MeOH/14.7m NH4OH, 99:1:0.1
Acknowledgements
The authors gratefully acknowledge Dr. Joel Krauser, Dr. Thomas
Moenius, and Dr. Albrecht Glꢀnzel for the radiosyntheses and
quality control of [3H]1 and [3H]33b; Dr. Esther Van de Kerkhof
for the autoradiographic analysis of [3H]33b, Carol Ginsberg-
Moraff for CHI(IAM) measurements, and Rrezarta Ramqaj, Ema-
nuele Mauro, and James Ternois for skillful synthetic support.
1
to 90:10:1) to give 33a as a white powder (34 mg, 49.2%): H NMR
(400 MHz, CDCl3): d=7.15 (t, J=8.4 Hz, 1H), 7.08 (d, J=8.5 Hz, 2H),
6.91 (d, J=8.56 Hz, 2H), 6.82 (d, J=8.44 Hz, 2H), 4.51 (dt, J=12.07,
6.01 Hz, 1H), 3.91 (s, 2H), 3.51–3.56 (m, 2H), 2.47 (brt, J=6.17 Hz,
2H), 1.76–1.92 (m, 4H), 1.32 ppm (d, J=6.11 Hz, 6H); UPLC–MS:
tR =0.88 min, m/z: 341.0 [M+H].
Keywords: imaging agents
·
inhibitor design
· ligand
efficiency · medicinal chemistry · prostacyclin receptor
N-(3-Fluoro-4-(4-(4-fluorobutoxy)benzyl)phenyl)-3,4,5,6-tetrahy-
dropyridin-2-amine (33b): Compound 33b was prepared from 4a
(200 mg, 0.51 mmol), according to the procedure described for the
[1] M. M. Hann, G. M. Keserꢃ, Nat. Rev. Drug Discovery 2012, 11, 355–365,
and references therein.
[2] a) T. T. Wager, R. Y. Chandrasekaran, X. Hou, M. D. Troutman, P. R. Verho-
est, A. Villabolos, Y. Will, ACS Chem. Neurosci. 2010, 1, 420–434; b) T. T.
Wager, X. Hou, P. R. Verhoest, A. Villabolos, ACS Chem. Neurosci. 2010, 1,
435–449; c) L. Zhang, A. Villalobos, E. M. Beck, T. Bocan, T. A. Chappie,
L. Chen, S. Grimwood, S. D. Heck, C. J. Helal, X. Hou, J. M. Humphrey, J.
Lu, M. B. Skaddan, T. J. McCarthy, P. R. Verhoest, T. T. Wager, K. Zasadny,
J. Med. Chem. 2013, 56, 4568–4579; d) Q. Guo, M. Brady, R. N. Gunn, J.
Nucl. Med. 2009, 50, 1715–1723.
1
synthesis of 33a (183 mg, 96%): H NMR (400 MHz, CDCl3): d=7.14
(d, J=8.44 Hz, 2H), 7.03 (t, J=8.25 Hz, 1H), 6.83 (d, J=8.68 Hz,
2H), 6.56–6.65 (m, 2H), 4.52 (dt, J=48 Hz, 5.56 Hz, 2H), 3.99 (t, J=
5.75 Hz, 2H), 3.88 (s, 2H), 3.23 (brs, 2H), 2.45 (brs, 2H), 1.69–
1.94 ppm (m, 8H); 13C NMR (151 MHz, CDCl3): d=165.9, 160.7,
156.6, 133.7, 131.8, 130.6, 129.8 (2C), 129.3, 122.0, 116.0 (2C), 113.6,
83.9 (d, J=165 Hz, 1C), 67.3, 42.1, 33.7, 27.3, 26.3, 25.2, 20.7,
18.1 ppm; UPLC–MS: tR =0.85 min, m/z: 373.7 [M+H].
[3] Z. Jiang, J. Reilly, B. Everatt, E. Briard, J. Pharm. Biomed. Anal. 2011, 54,
722–729.
[4] K. Valko, C. M. Du, C. D. Bevan, D. P. Reynolds, M. H. Abraham, J. Pharm.
Sci. 2000, 89, 1085–1096.
Pharmacology
[5] M. D. Shultz, Bioorg. Med. Chem. Lett. 2013, 23, 5980–5991.
[6] a) F. Assmus, A. Seelig, L. Gobbi, E. Borroni, P. Glaentzlin, H. Fischer, Eur.
J. Pharm. Sci. 2015, 79, 27–35; b) S. Belli, F. Assmus, B. Wagner, M.
Honer, H. Fischer, F. Schiuler, R. Alvarez-Sanchez, Mol. Pharm. 2015, 12,
4529–4541.
[7] J. Mukherjee, Z. Y. Yang, M. K. Das, T. Brown, Nucl. Med. Biol. 1995, 22,
283–296; J. Mukherjee, Z. Y. Yang, T. Brown, R. Lew, M. Wernick, X.
Ouyang, N. Yasillo, C. T. Chen, R. Mintzer, M. Cooper, Nucl. Med. Biol.
1999, 26, 519–527.
Saturation binding assays: Binding was performed with a range of
concentrations of [3H]1 to construct saturation binding curves, as
described previously.[23] Experiments were conducted at room tem-
perature in 96-well deep-well plates in assay binding buffer (HBSS
containing 5 mm HEPES, 0.02% pluronic acid, and 2% DMSO,
pH 7.4). Membranes of Chinese hamster ovary (CHO) cells express-
ing hIPR (2 mg per well) were incubated in 96-well deep-well plates
&
ChemMedChem 2016, 11, 1 – 14
www.chemmedchem.org
12
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
[8] M. Slifstein, D. R. Hwang, Y. Huang, N. Guo, Y. Sudo, R. Narendran, P.
Talbot, M. Laruelle, Psychopharmacology 2004, 175, 274–286.
[9] J. R. Bunzow, G. Zhang, C. Bouvier, C. Saez, O. K. Ronnekleiv, M. J. Kelly,
D. K. Grandy, J. Neurochem. 1995, 64, 14–24.
[10] H. Sugimoto, Y. Iimura, Y. Yamanishi, K. Yamatsu, J. Med. Chem. 1995, 38,
4821–4829.
[18] G. P. Donahich, A. E. Johnson, B. Nock, B. S. McEwen, H. H. Feder, Eur. J.
Pharmacol. 1985, 119, 9–16.
[19] S. Hintermann, I. Vranesic, H. Allgeier, A. Brꢃlisauer, D. Hoyer, M. Lem-
aire, T. Moenius, S. Urwyler, S. Whitebread, F. Gasparini, Y. P. Auberson,
Bioorg. Med. Chem. 2007, 15, 903–914.
[20] K. R. Bley, R. D. Clark, A. Jahangir, B. A. Kowalczyk, F. J. Lopez-Tapala,
A. V. Muehldorf, C. O’Yang, T. W. Sun (F. Hoffmann–La Roche AG), Eur.
Pat. No. EP0902018A3, 1999.
[11] H. Sugimoto, Y. Yamanishi, Y. Iimura, Y. Kawakami, Current Med. Chem.
2000, 7, 303–339.
[12] B. Giros, S. el Mestikawy, N. Godinot, K. Zheng, H. Han, T. Yang-Feng,
M. G. Caron, Mol. Pharmacol. 1992, 42, 383–390.
[21] M. Nonnenmacher, T. Busch (Euticals S.P.A.), Int. PCT Pub. No.
WO 2014180735 A1, 2014.
[13] C. V. Cella, S. M. Woodhouse, R. A. Lewis, A. E. Fray, Abstracts: 4th Forum
of European Neuroscience, Lisbon (Portugal), 2004, 2, A059.4.
[14] M. Schou, V. W. Pike, C. Halldin, Current Top. Med. Chem. 2007, 7, 1806–
1816.
[15] P. Kannan, V. W. Pike, C. Halldin, O. Langer, M. M. Gottesman, R. B. Innis,
M. D. Hall, Mol. Pharm. 2013, 10, 2222–2229.
[16] M. Healy, Abstracts: 250th National Meeting of the American Chemical So-
ciety, Boston, MA (USA), August 16–20, 2015, ACS, Washington DC,
MEDI 18.
[22] R. Frank, T. Christoph, K. Schiene, V. J. De, N. Damann, B. Lesch, G. Bah-
renberg, D. J. Saunders, H. Stockhausen, Y.-S. Kim, M.-S. Kim, J. Lee (Grꢃ-
nenthal GmbH), Int. PCT Pub. No. WO2013068461, 2013.
[23] D. A. Sykes, M. R. Dowling, S. J. Charlton, Curr. Protoc. Pharmacol. 2010,
50, 9.14.1–9.14.30.
[24] C. M. Carter, J. R. Leighton-Davies, S. J. Charlton, J. Biomol. Screening
2007, 12, 255–266.
Received: February 23, 2016
Revised: April 13, 2016
[17] R. D. Clark, A. Jahangir, D. Severance, R. Salazar, T. Chang, D. Chang,
M. F. Jett, S. Smith, K. Bley, Bioorg. Med. Chem. Lett. 2004, 14, 1053–
1056.
Published online on && &&, 0000
ChemMedChem 2016, 11, 1 – 14
www.chemmedchem.org
13
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

FULL PAPERS
Y. P. Auberson,* E. Briard, D. Sykes,
J. Reilly, M. Healy
Striking a balance: Nonspecific binding
cannot be predicted using in silico pa-
rameters and remains a major cause of
failure for candidate PET imaging
&& – &&
Ligand Specific Efficiency (LSE) Index
for PET Tracer Optimization
agents. We defined the ligand specific
efficiency (LSE) index as a measure of af-
finity, normalized to nonspecific bind-
ing, and determined the minimal value
for a successful PET tracer. The use of
LSE to guide chemical optimization is il-
lustrated with tracer candidates for
prostacyclin receptors.
&
ChemMedChem 2016, 11, 1 – 14
www.chemmedchem.org
14
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!