2
H.-Y. Lee et al. / Bioorg. Med. Chem. Lett. xxx (2016) xxx–xxx
Our compounds were not potent against Gram-negative bac-
teria (MIC > 50 g/mL) in minimum inhibitory concentration
l
OH
OH
N
O
(MIC) assay. Possible explanations may include factors such as
poor cell penetration or efflux pump of the bacteria. In the case
of Gram-positive bacteria, aminopiperidines with unsubstituted
or substituted five-membered pyrrole and furan group did not
show antibacterial activity, with the exception of PRSP (9a, 9e,
9f), but six-membered aromatic group (9b, 9c) showed results
similar to that of vancomycin and linezolid. In contrast, the
six-membered pyridine group (9d) was proven to be not an
effective bacterial inhibitor compared to 9b, since its hydropho-
bic deficiency had a negative effect on its potency. The antibac-
terial effect of the bond length of the linker between
aminopiperidine and (hetero)aryl group did not affect potency
(9b, 9c, 9g, 9h), and other fused heteroaryl ring compounds
did not show noticeable in vitro activity comparable to linezolid
and vancomycin (9i, 9j, 9k). Consequently, six-membered aro-
matic substitutes 9b and 9c were similar to linezolid and van-
comycin in terms of antibacterial activity. Based on our
observation of the physical properties of 9b and 9c, benzyl group
(9b) was more soluble than phenyl group (9c), so we selected 9b
for further SAR studies.
We have evaluated benzylaminopiperidine derivatives of com-
pound 9b since it may demonstrate different antibacterial activity
when there is one or more substituent (Table 2). For improved
antibacterial activity, we explored the electronic effects of func-
tional group N-formylhydroxylamine. Assessing from the results
shown in Table 2, electronic effects were not important, but it
was apparent that proton donor groups completely lost their
antibacterial activity. For example, antibacterial activity of 10a
was equivalent to that of 10e, but proton donor groups, such as
hydroxy (10c, 10h) and amide (10g) did not display any activity
against MRSA and VRE. Non-proton substituent groups (10a, 10b,
10d, 10e, 10f, 10i, 10j) showed good antibacterial activity
O
O
H
N
H
N
O
HO
O
O
N
N
H
N
H
O
N
O
Actinonin
BB-83698
OH
N
N
H
N
OH
N
F
O
N
O
N
N
H
N
H
N+
O-
H
O
O
H
O
F
N
O
N
H
Lanopepden
LBM-415
Figure 1. Representative PDF inhibitors.
SAR of P30, we introduced aminopiperidine substituents, such as
five- or six-membered or bicyclic (hetero) aromatic group.
The general method for synthesis of the aminopiperidine moi-
ety is described in Scheme 1. Reductive amination of the starting
materials N-Boc-4-aminopiperidine or N-Boc-4-piperidone (1)
and corresponding carboxaldehyde or amine by sodium triacetoxy-
borohydride were given to N-Boc-4-N-substituted aminopiperidine
(2). The protection of 2 using benzyloxycarbonyl group under basic
condition, followed by the deprotection of 3 under acidic condition,
yielded amine hydrochloride intermediate 4. For P20 moiety, reac-
tion of 4 with Boc-
intermediate 5, through 4-(dimethylamino)pyridine (DMAP) and
1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride
(EDCI). Subsequently, the removal of the protecting group under
acidic condition yielded key intermediate (hetero) ary-
laminopiperidine hydrochloride 6. All intermediates displayed
higher yield and purity via solidification or acid/base workup than
via column chromatography.
L-tert-leucine was performed to yield an amide
(MIC < 1 lg/mL) against PRSP. Compounds 10b, 10f and 10i were
The N-formylhydroxylamine intermediate 7 was synthesized
from commercially available acid through a sequence of reactions
which were described in previous studies.12 Intermediate 8 was
prepared from (hetero) arylaminopiperidine hydrochloride 6 and
intermediate 7 via amide coupling conditions without purification.
Finally, through the removal of the protection group by palladium-
catalyzed hydrogenolysis, we obtained quantitative and optically
pure N-formylhydroxylamine 9a–9k or 10a–10j.
The antibacterial activity of these (hetero) arylaminopiperidine
derivatives against selected resistant Gram-positive (MRSA, VRE
(faecalis, faecium), PRSP) and Gram-negative pathogens (E. coli,
Pseudomonas aeruginosa) (P. aeruginosa) were evaluated in compar-
ison to that of linezolid and vancomycin. The in vitro results are
summarized in Table 1.
less active than unsubstituted 9b when tested against all of the
pathogens, but 10d was twice as more potent than linezolid
against VRE (faecium) and PRSP. Antibacterial activity of tri-substi-
tuted 10j was higher than that of mono-substituted 10e against
MRSA and VRE (faecalis), and this supports the idea that the greater
the compound’s lipophilic structures are, the more permeable they
are to bacterial cell. Compounds 10d and 10j were more potent
than any other derivatives against Gram-positive pathogens, and
10j especially showed potency equal to or higher than that of
linezolid.
As a result, antibacterial activities were observed in compounds
10d and 10j. Additionally, enzyme inhibitory test of these com-
pounds against S. aureus PDF was evaluated according to the liter-
ature.13 The half maximal inhibitory concentration (IC50) values
were similar to that of BB-83698, we were able to identify amino-
piperidine derivatives 10d and 10j to have excellent antibacterial
properties (Table 3).
S1'
The pharmacokinetics of 10d and 10j was studied using mice
models. Our amidopiperidine derivatives were previously con-
firmed to have shown a low pharmacokinetic profile. As shown
in Table 4, for amidopiperidine derivatives having the same sub-
stituents as 10d, although they exhibit similar clearance, a low
therapeutic efficacy in in vivo animal models is expected as a result
of low oral absorption. On the other hand, 10d and 10j displayed
high absolute oral bioavailability (%F) in mice models, with rapid
absorption of these compounds after oral administration. In the
case of intravenous administration, 10d was quickly eliminated,
with total plasma clearance (CL) of 3.09 L/h/kg, with an apparent
terminal elimination half-life (t1/2) of 0.36 h. However, 10j showed
a 2.71-fold lower clearance and 2.91-fold longer half-life when
compared to 10d in mice.
Fe2+
P1'
O
S3'
H
N
MBG
P3'
N
H
O
P2'
S2'
Generic PDF Inhibitor
(MBG = Metal Binding Group)
OH
N
O
OH
N
O
H
N
H
N
O
O
N
N
O
H
O
H
O
N
H
R
N
H
R
Aminopiperidine derivatives
Amidopiperidine derivatives
Figure 2. Generic PDF inhibitor and aminopiperidine derivatives.
Lee, Hee-Yeol
