M. Yu et al. / Bioorg. Med. Chem. Lett. 24 (2014) 156–160
157
Cl
synthesized from 4-methyl-3-nitroaniline via Buchwald coupling
with A (R1 = –Me, R2 = –iPr), followed by hydrogenation. Treating
C2 with 4-chlorobenzenesulfonyl chloride in pyridine completed
the synthesis of 16 (Table 2) in 18% yield.
O
iPrO
N
N
N
N
O
S
O
S
O
O
N
N
iPr
N
O
O
N
H
O
N
We first investigated the influence of the pyridylphthalimide R1
and R2 substituents on potency as summarized in Table 1. Starting
with R2 = –cPr as like in 5, R1 groups in a wide range of size and
electronic properties were tested. Many of them were found to
be well tolerated for this position, including –Cl and –OCF3 (data
Cl
1
2
hGPR119-cAMP EC50 = 0.024 µM
hGPR119-cAMP EC50 = 0.041 µM
Figure 1. Representative synthetic GPR119 agonists.
not shown). From this exploration, the methoxy (6, 0.019
lM)
and methyl (7, 0.015 M) groups stood out as the most potent sub-
l
stituents. The methyl group was later selected over methoxy for R1
for the subsequent SAR, primarily due to its potential for improved
metabolic stability. The R2 position was also quite sensitive toward
substituent size. From compounds 7 to 11, the EC50 values were
ranked in order of low to high potency as R2 = –H > –Me > –cPr,
–iPr > –Et, but larger substituents (e.g. 12 and 13) led to significant
loss of potency. The combination of R1 = –Me and R2 = –iPr, as
shown in compound 11, offered a twofold improvement in potency
N
O
S
O
N
O
CO2H
H
CN
3
Cl
cPrNH2, EDC, DCM/DMF
rt, 12 hr
(EC50 = 0.017 lM) compared to 5. Compound 11 also demonstrated
in vitro metabolic stability which, while poor (see below), was still
slightly better than the more potent compound 7 and 10. There-
fore, the chemical feature of R1 = –Me and R2 = –iPr as in 11 was
adopted for subsequent studies.
We also explored modifications to the central –O– linker as
highlighted in Table 2. We first replaced it with –S– as exemplified
in 14 and 15, but this change either showed no benefit or was det-
N
O
O
N
O
S
O
S
N
O
O
N
O
CONHcPr
Cl
H
N
H
CN
O
Cl
4
5
(32%)
(15%)
Scheme 1. Discovery of 5 from the synthesis of 4.
rimental to potency (0.019 and 0.224
tested an –NH– substituent, as exemplified in 16, but this was also
detrimental to potency (0.382 M). In addition, both the –S– and –
lM, respectively). We then
–iPr; 31–51%), or utilizing a copper (II) acetate catalyzed coupling
reaction with the relavent boronic acids R2B(OH)2 (R2 = –cPr or –
Ph; 11–68%).
l
NH– replacements reduced the efficacy by approximately 20%.
As shown in Scheme 3, another class of key building compo-
nents for most of the compounds discussed in this Letter were
the N-(3-hydroxyphenyl)-benzenesulfonamides B1. They were
prepared in 25–87% yields by reaction of 4-chlorobenzenesulfonyl
chloride with the corresponding 3-aminophenols of required sub-
stitution pattern (Z = –H, 2-F, 4-Me, 5-F, 6-F, or 6-Cl).8 Intermediate
B1 (Z = H) was initially assembled through a potassium carbonate-
mediated SN2-Ar reaction with A (R1 = R2 = –cPr) to provide 5 (Ta-
ble 1), and with A4 (R1 = –Me) to provide 8, in 5% and 3% yield
respectively. Compounds 17 and 18 (Table 3) were similarly
assembled using the corresponding B2 and B3 intermediates made
through procedures outlined in Scheme 3. Since this protocol was
generally low yielding (3–12%), a 1,4-diazabicyclo[2.2.2]octane
(DABCO) catalyzed microwave assisted reaction (DMF, 130 °C)
was developed which improved the reaction yields to 41–52% for
the synthesis of compounds 5–7, 9–13 (Table 1) and 20–24
(Table 4).10 Compound 19 (Table 4) was made through direct
N-methylation of 11 with MeI in 72% yield. The synthesis of
14–15 and 16 (Table 2) started with construction of the right hand
side C1 or C2 components of the molecule. Subjecting 3-amino-
thiophenols to conditions (K2CO3, THF, 60 °C) similar to, but milder
than those described above provided C1 in 99% (Z = –H) and 80%
(Z = –Me) yields. Treating C1 with 4-chlorobenzenesulfonyl
chloride in pyridine gave 14 and 15 in 99% and 90% yields.
The aminoaniline–pyridylphthalimide intermediate C2 was
Consequently, no further efforts were made in this area.
Pharmacokinetic studies of 11 suggested rapid metabolic
decomposition of this compound in both rat liver microsomes
(RLM) (CLintrinsic = 250 l
l/min/mg)11 and in vivo (rat IV CL = 4.8 L/
h/kg). Oxidative metabolism of the central phenyl ring was identi-
fied as the major route of metabolism for 11 in RLM. In order to im-
prove the metabolic stability we reduced the electron density on
the central phenyl ring, as illustrated by the examples shown in Ta-
ble 3. In 17, the connectivities of the sulfonamide moiety were re-
versed to reduce the electron density and therefore to lower the
oxidative potential of the central phenyl ring. However, this com-
pound displayed over 20-fold loss of potency (0.238
where a carboxylamide served as a sulfonamide surrogate, there
was a noticeable improvement in RLM intrinsic clearance (40 l/
min/mg). Unfortunately, compound 18 showed no activity in the
functional assay (EC50 >30 M).
lM). In 18,
l
l
Given the modest improvement achieved from the efforts
described above, and our knowledge from previous SAR8 that the
4-Cl phenyl motif at the far left side is a well established chemical
feature of this lead series, we decided to revisit the central phenyl
ring with more delicate modifications (Table 4). To better manipu-
late the electron density around the central phenyl ring to increase
its oxidative potential, methyl substitution was introduced at the
sulfonamide nitrogen (19, Y = –Me, Z = –H) to disrupt the electron
donating conjugation effect from the electron lone pair of the nitro-
R1
R1
R1
R1
a
b
c
d or e or f
O
O
N
N
N
N
R1
OEt
Cl
Cl
O
O
HO
CO2Et
Cl
CO2Et
N
N
CN
A2
CN
A3
O
R2
O
H
(R1 = Me, cPr)
A4
A
A1
Scheme 2. Synthesis of key pyridylphthalimide intermediate A. Reagents and conditions: (a) 2-cyanoacetamide, piperidine, EtOH, 80 °C, 35–59%; (b) PhP(O)Cl2, 150 °C, 25–
45 min, 55–76%; (c) H2SO4 (6.0 M), 80 °C, 4.0–8.0 h, 56–72%; (d) R2Br, Cs2CO3, DMF, rt , 42–55%; (e) R2OH, DEAD, Ph3P, THF, 0 °C–rt, overnight, 31–51%; (f) R2B(OH)2, Cu(OAc)2,
2,20-dipyridine, Na2CO3, DMF, 70 °C, 1.5–4.0 h, 11–68%.