X.-L. Liu et al. / Bioorg. Med. Chem. Lett. 26 (2016) 4698–4701
4699
S
O
R2
NH
O
O
O
S
CO2H
S
CO2H
S
CO2H
N
H
S
CO2H
S
R1
N
H
N
H
S
N
H
S
O
O
O
O
A
B
C
Amino acid thioester scaffold
IC50=0.018uM, Ki=0.11uM
IC50=2.9uM, Ki=0.95uM
IC50=0.29uM, Ki=0.17uM
Figure 1. Amino acid thioester scaffold and structures of specific compounds A–C reported previously.18
10 lacking an amino acid side chain was also included as a
reference.
nyl methyl is larger than the benzyl, and the benzyl is larger than
the pentyl group. Within the series with a constant 4-diphenyl
methyl R1 group, 1 and 7 are more potent than 4, while in the ser-
ies with a constant benzyl R1 group, 2 and 8 are more potent than
5, implying that the tryptophan and methionine amino acid side
chains (R2) are more favorable than the phenylalanine side chain,
which is in agreement with previous results.18 With the aliphatic
pentyl R1 group, the tryptophan side chain as R2 in 3 did not result
in higher potency than the phenylalanine side chain in 6. 9 had
almost identical potency as 8, indicating that the phenyl R1 group
is nearly equivalent to the benzyl R1 group. Furthermore, com-
pared with the previously reported compounds B and C,18 the inhi-
bitory activities of 1 and 4 are increased, indicating that the 4-
diphenyl methyl R1 group is superior to the 2-thiophenyl methyl
group.
The synthetic pathway of the ten thioesters is shown in
Scheme 1. Firstly, hexanoic acid, phenylacetic acid, 4-bipheny-
lacetic acid, and benzoic acid, respectively, were refluxed with sul-
fur dichloride to get the corresponding acyl chlorides.19,20
Secondly, racemic mixtures of amino acids reacted respectively
with the four kinds of acyl chloride to give amides 1b–9b. Finally,
the amides reacted with mercaptoacetic acid to give the target
products mercaptoacetic acid thioesters 1–9. The synthetic process
of 10b and 10 is similar to that of 1b–9b and 1–9. The structures of
the synthesized mercaptoacetic acid thioesters are shown in Fig-
ure 2. These synthesized thioesters were all characterized by 1H
and 13C NMR and confirmed by MS (see Supporting information).
Due to racemic mixtures of amino acids being used, the products
are not expected to be chiral.
To test whether the inhibitory activities of these mercaptoacetic
acid thioesters against L1 have improved, L1 was over-expressed
and purified as previously described.21 In vitro, the inhibitory
activities of all compounds prepared were tested against L1 on
an Agilent UV8453 spectrometer as described by Bush et al. using
cefazolin as the substrate.22 The substrate concentration was
The capacity of the mercaptoacetic acid thioesters to restore the
antibacterial activity of cefazolin against Escherichia coli cells
expressing L1 was investigated by determining the minimum inhi-
bitory concentrations (MICs) of cefazolin in the absence and pres-
ence of 16 lg/mL 1–10. MIC values were determined by using the
Clinical and Laboratory Standards Institute (CLSI) macrodilution
(tube) broth method.23 The bacterial strain of E. coli BL21(DE3)
containing plasmids pET26b-L1, as well as an E. coli BL21(DE3)
control without plasmid, were used to assess these inhibitors.
60
lM, and inhibitor concentrations were varied between 0.01
and 1
l
M. Enzyme and inhibitor were pre-incubated for 30 min
before starting the kinetic assays. The inhibitor concentrations
causing 50% decrease of enzyme activity (IC50) were calculated
based on the kinetic data. The IC50 values of the ten compounds
against L1 with cefazolin as substrate are listed in Table 1. It can
be observed that the thioesters 1–9 exhibited strong inhibition of
The final concentration of inhibitor was 16 lg/mL. The data listed
in Table 2 shows that inhibitors 1–9 resulted in at least 4-fold
reduction of MIC for E. coli BL21(DE3) expressing L1, effectively
restoring the MIC observed with E. coli not expressing L1, indicat-
ing successful inhibition of L1 in vivo (inside the bacterial cells).
Inhibitor 10 did not decrease the MIC relative to the blank control,
indicating no inhibition of L1, which is consistent with the absence
of any inhibition in vitro.
L1 with an IC50 value range of 0.02–0.6 lM, while compound 10
had no inhibitory activity. A possible reason is that the amino acid
in its structure was replaced by 4-aminobenzoic acid. Clearly, the
IC50 values of 1, 4 and 7 are smaller than those of 2, 5 and 8, respec-
tively, revealing that the 4-diphenyl methyl as R1 results in stron-
ger inhibitory activity against L1 than the benzyl. IC50 values of 2
and 5 are lower than the corresponding values of 3 and 6, respec-
tively, indicating that the aromatic R1 improves inhibitory activity
compared to the aliphatic pentyl group. This activity relationship is
related to the space sizes of these substitutes, that is, the 4-diphe-
In order to clarify why the introduction of 4-biphenyl methyl as
R1 can enhance inhibitory activity, 4, 5 and 6 were docked into the
active site of the L1 crystal structure (PDB code 2AIO)24 using the
same procedure as reported previously.25 We chose these com-
pounds because of a significant improvement of their inhibitory
effect compared to inhibitor C (IC50 = 2.9
formations (the top ranked conformations) of 4, 5, and 6 docked
l
M).18 Low-energy con-
R2
O
R2
O
R2
H2N
CO2H
c
b
S
CO2H
CO2H
R1
N
H
CO2H
R1
N
H
O
O
a
O
R1 Cl
1-9
R1 OH
1b-9b
O
d
e
Cbz
HN
S
H2N
CO2H
CO2H
Cbz
,
N
H
10b
10
R1=
NH
R2
(CH2)4
,
,
,
S
,
Scheme 1. Synthetic route of mercaptoacetic acid thioesters. Reagents and conditions: (a) SOCl2, DMF, 80 °C, 3 h; (b) NaOH, diethyl ether, HCl; (c) mercaptoacetic acid, ethyl
chloroformate, Et3N, AcOEt/DMF; (d) CbzCl, NaOH, diethyl ether, HCl; (e) mercaptoacetic acid, ethyl chloroformate, Et3N, DMF.