Journal of the American Chemical Society
M/B = 50) was added to the reaction until it was also
Communication
(
pathogenic Gram-positive bacterium responsible for severe
34,35
consumed (FTIR monitoring). The resulting poly(sar) 9
hospital acquired intestinal infections: C. difficile
and
−
1
presented a M of 3200 g·mol close to the M of polymer 3
copolymers 11−21 were evaluated in vitro against this
pathogen (Table 2). In this study, the bacteria were incubated
for 24 h with increasing concentrations of cyclic polypeptoids,
to determine the minimum inhibitory concentrations (MICs).
In parallel, the in vitro cytotoxicity of the best polymers (15,
19−21) was tested on a human intestinal epithelial cell line
(Caco-2), allowing us to measure the 50% cytotoxic
concentrations (CC ) and to calculate the selectivity indices
w
w
and associated with a low dispersity of Đ = 1.1 (Table 1). This
result indicated that diblocks could be designed using
LiHMDS.
Cyclic structures are used by living systems to achieve
27
specific bioactivities. This encompasses cyclic antimicrobial
peptides (cAMP) that are cationic and broad-spectrum anti-
28
infective backbones. The use of a simple REP methodology,
working in DMF, opens up interesting prospects to design
simplified analogues of such cAMPs. Up to now, only a few
monomers have been used in REP because of some limitations
related to the solvents used (in which, for example,
5
0
(SI = CC /MIC). Optimizing the cationic content of
5
0
copolymers was a key design to optimize the antibacterial
activity and copolymer 15 showed a good profile, similar to the
(Table S7). Interestingly, this statistical copolymer 15 was
significantly less toxic than copolymer diblock counterparts
15 by propylation, Figure S26). We also observed that
marked contrast to LL37 (Figure S28). This resistance to
proteolysis and the ability to destabilize phospholipid
liposomes (Figure S29). Overall, this proof-of-concept
demonstration shows that LiHMDS-mediated REP enables
quick and simple access to libraries of copolymers to design
anti-infective backbones.
29,30
poly(sarcosine) was not soluble).
Considering that both
the hydrophobic and cationic characters are key structural
31
features to design antimicrobial polymers, we consequently
prepared copolymers by mixing N-protected-lysine-like NNCA
(
Z-N-lys-NCA) and phenylalanine-like NNCA (N-phe-NCA).
mediated REP of Z-N-lys-NCA was carried out in DMF at M/
LiHMDS = 50, by itself (polymer 10 Table 2 and Supporting
Table 2. In Vitro Antibacterial Activity against C. difficile and
Cytotoxicity over Caco-2 Cell Line
In summary, the use of a LiHMDS-mediated REP of NNCA
in DMF was a versatile and facile route to cyclic polypeptoids
allowing fine-tuning over their chemical composition. This new
methodology permits us to design cationic macrocycles
showing good activity against C. difficile, the main causative
agent of nosocomial diarrheas of adults in developed countries.
Overall, the use of LiHMDS is a simple way to design original
protease-resistant backbones with proper bioactivity.
cationic
content
C. diff. MIC Caco-2 CC50
a
b
c
polymer (M/B = 50)
(%)
(μg/mL)
(μg/mL)
SI
d
P(N-lys) 10
100
>100
>100
>100
>100
d
P(N-lys-r-N-phe) 11
P(N-lys-r-N-phe) 12
P(N-lys-r-N-phe) 13
P(N-lys-r-N-phe) 14
P(N-lys-r-N-phe) 15
93 (90)
85 (80)
69 (70)
61 (60)
50 (50)
50 (50)
95 (90)
79 (80)
68 (70)
59 (60)
49 (50)
52 (50)
12.5
d
d
d
>100
12.5
12.5
>100
>100
>100
208 ± 9
17
1
5 post-trypsin
ASSOCIATED CONTENT
sı Supporting Information
■
d
P(N-lys-b-N-phe) 16
P(N-lys-b-N-phe) 17
P(N-lys-b-N-phe) 18
P(N-lys-b-N-phe) 19
P(N-lys-b-N-phe) 20
P(N-lys-r-N-phe) 21
*
d
d
12.5
100
25
213 ± 7
77 ± 6
105 ± 8
47 ± 1
17
6
<1
2
Synthetic procedures, macromolecular characterizations
(NMR, ESI-MS, FTIR, SEC, MALDI-ToF), sample
preparations, spectroscopic analyses, molecular model-
ing and biological assays, Cartesian coordinates, plot of
e
LL37
d
LL37 post-trypsin
doxorubicin
>200
f
8.1 ± 0.1
a
b
1
r = random copolymers and b = block copolymers. From H NMR
c
d
(
in bracket, targeted). Selectivity index = CC /MIC. The MIC
50
e
value was not reached at the highest tested concentration. LL37 was
f
used as AMP reference. Doxorubicin was used as a cytotoxic drug
■
control.
Corresponding Authors
Pierre Verhaeghe − LCC-CNRS, UPR8241, Université de
Colin Bonduelle − Université Bordeaux, CNRS, Bordeaux
as a one-pot synthesis (leading to statistical copolymers 11−
1
5, Table 2) or in sequence (leading to block copolymers 16−
0, Table 2). For all copolymers, acidic deprotection of the
2
Cbz group (HBr 2 equiv in TFA) afforded side-chain
deprotection, allowing producing cationic macrocycles show-
Authors
The bioactivity of the copolymers was then assessed. Cyclic
peptoids are promising scaffolds to design anti-infective
Pedro Salas-Ambrosio − Université Bordeaux, CNRS,
Bordeaux INP, LCPO, UMR 5629, F-33600 Pessac, France
Antoine Tronnet − LCC-CNRS, UPR8241, Université de
Toulouse, CNRS, UPS, 31400 Toulouse, France; LPBA,
3
2,33
agents,
but antibacterial copolymers have never been
developed on the basis of this chemical design. We targeted a
3
700
J. Am. Chem. Soc. 2021, 143, 3697−3702