A. Pasternak et al. / Bioorg. Med. Chem. Lett. 18 (2008) 994–998
995
shown to be the best. This information guided us in the
selection of our initial aryl and heteroaryl piperidine
containing analogs. The requisite aryl and heteroaryl
piperidines were prepared according to the procedure
of Wustrow and Wise by palladium-mediated coupling
of aryl and heteroarylboronic acids with tert-butyl 4-
{[(trifluoromethyl)sulfonyl]oxy}-3,6-dihydropyridine-
1(2H)-carboxylate, followed by hydrogenation and Boc
removal.10 The resulting piperidines were reductively
alkylated with previously described7b intermediates 2
and 3 by treatment with sodium triacetoxyborohydride
(Scheme 1) over 2–3 days. The cis and trans-isomers
could usually be separated by preparative TLC (with
the cis-isomers normally being obtained as the major
isomers). Slow reductive amination times were tolerated
because of good yields and favorable cis-selectivities;
alternative methods were not evaluated. The two cis-dia-
stereomers could then be separated by preparative chiral
HPLC (using either a Chiralcel OD or a Chiralpak AD
column, obtained from Chiral Technologies, Inc.) to
afford the homochiral analogs. In some cases functional
groups on the phenyl ring were modified to give the final
analogs (i.e., esters hydrolyzed).
and 3-fluoro (4a) direct analogs. Similarly, introduction
of a 3-pyridyl (4f) in place of the phenyl (1a) gave only a
slight improvement in selectivity, although potency
remained high at CCR2 indicating a tolerance for
heteroaryl groups at this position. The largest decrease
in IKr affinity was observed upon introduction of a car-
boxylic acid in the 3-position (4d), however this was
accompanied by a simultaneous decrease in CCR2 bind-
ing affinity. The effect was more pronounced in the
hydroxypropyl side chain series (4e) where IKr binding
dropped to 20 lM, while CCR2 binding, though dimin-
ished, remained at a respectable 45 nM.
The modest selectivity enhancement and the CCR2
potency retention observed with 3-pyridyl analog 4f
prompted us to prepare more polar heteroarylpiperidine
containing analogs. In particular, we were interested in
preparing all possible two nitrogen containing 6-mem-
bered heteroaryl isomers, as these would be more polar
than the pyridylpiperidine but of similar steric dimen-
sions. At the time of this work we could find no litera-
ture reference to the target pyrimidyl, pyridazyl, and
pyrazyl piperidines. Since the heteroarylboronic acids
were not available, we decided to devise a synthesis that
could make use of commercially available haloheteroa-
ryl precursors. Recently Billotte, in a letter focusing
on aryl azetidine synthesis from an azetidinyl zincate
reagent, briefly described the application of the piperid-
inyl zincate 6 to the synthesis of two aryl piperidines via
palladium mediated coupling to the corresponding aryl
halides.12 As shown in Scheme 2 and Table 2, we suc-
cessfully applied Billotte’s methodology to the synthesis
of all of our target isomeric heteroarylpiperidines, lead-
ing to the first disclosure of a general preparation of
these interesting and potentially useful new compounds.
4-Substituted piperidines prepared according to Scheme
2 were reductively alkylated with 2 and 3 as described in
Scheme 1 to give, after separation of the isomers, target
analogs for evaluation in our assays. Some of the hetero-
arylpiperidines were prepared with di or trihalo hetero-
aryl precursors. The resulting haloheteroarylpiperidines
could be hydrogenated prior to Boc removal (Scheme 2,
step d and 8e) or they could be carried on to target
CCR2 antagonists according to Scheme 1 and then
hydrogenated so that both the haloheteroaryl and re-
duced heteroarylpiperidines could be evaluated in our
assays.
CCR2 Binding affinities were determined using a radio-
ligand competition binding assay measuring inhibition
of 125I-MCP-1 binding to the endogenous CCR-2
receptor on human monocyte whole cells.7 IKr binding
data were obtained by measuring displacement of 33S-
MK499 from HEK cells stably expressing hERG.11
Functional blockade of IKr was not evaluated for these
analogs. An initial series of analogs with corresponding
CCR2 and IKr binding data is presented in Table 1.
Introduction of methoxy and hydroxy groups in the
phenyl 3-position provided only modest improvement
in selectivity as measured by the ratio of IKr binding
to CCR2 binding when compared to the 4-fluoro (1a)
O
O
CF3
N
H
R
CF3
2, R = H
3, R = OH
Ar
NaB(OAc)3H, Et3N,
A sieves, DCM
NH
4
º
Ar
Ar
O
R
Table 3 lists the target heteroarylpiperidine containing
analogs prepared, along with corresponding binding
data. Only analogs deemed of sufficient potency in our
CCR2 binding assay were evaluated in the IKr assay.
Most analogs were assayed as a mixture of two cis-dia-
stereomers, although we know from previously de-
scribed work that only the 1S,3R-isomer (as shown)
contributes to CCR2 binding affinity.7 All heteroarylpi-
peridine containing analogs except 4h, 4i, and 4j were
significantly less potent than the original lead 1 as well
as the pyridylpiperidine analog 4f. Compounds 4h, 4i,
and 4j, all containing the same 5-(pyrimidyl)piperidine
subunit, were of similar potency to lead 1 and pyridylpi-
peridine 4f. Compounds 4i and 4j are single 1S,
3R-enantiomers obtained by preparative chiral HPLC
N
CF3
N
H
CF3
Preparative TLC and/or
chiral preparative HPLC
O
N
CF3
N
H
R
CF3
4
Scheme 1. Synthesis of target analogs 4.
Pasternak, Alexander
