Synthesis and bioactivities study of new antibacterial peptide mimics: The dialkyl cationic amphiphiles

Article abstract of DOI:10.1016/j.ejmech.2017.10.044

The emergence of infectious diseases caused by pathogenic bacteria is widespread. Therefore, it is urgently required to enhance the development of novel antimicrobial agents with high antibacterial activity and low cytotoxicity. A series of novel dialkyl cationic amphiphiles bearing two identical length lipophilic alkyl chains and one non-peptidic amide bond were synthesized and tested for antimicrobial activities against both Gram-positive and Gram-negative bacteria. Particular compounds synthesized showed excellent antibacterial activity toward drug-sensitive bacteria such as S. aureus, E. faecalis, E. coli and S. enterica, and clinical isolates of drug-resistant species such as methicillin-resistant S. aureus (MRSA), KPC-producing and NDM-1-producing carbapenem-resistant Enterobacteriaceae (CRE). For example, the MIC values of the best compound 4g ranged from 0.5 to 2 μg/mL against all these strains. Moreover, these small molecules acted rapidly as bactericidal agents, and functioned primarily by permeabilization and depolarization of bacterial membranes. Importantly, these compounds were difficult to induce bacterial resistance and can potentially combat drug-resistant bacteria. Thus, these compounds can be developed into a new class of antibacterial peptide mimics against Gram-positive and Gram-negative bacteria, including drug-resistant bacterial strains.

Full text of DOI:10.1016/j.ejmech.2017.10.044

Accepted Manuscript
Synthesis and bioactivities study of new antibacterial peptide mimics: The dialkyl
cationic amphiphiles
En Zhang, Peng-Yan Bai, De-Yun Cui, Wen-Chao Chu, Yong-Gang Hua, Qin Liu,
Hai-Yang Yin, Yong-Jie Zhang, Shangshang Qin, Hong-Min Liu
PII:
S0223-5234(17)30844-9
10.1016/j.ejmech.2017.10.044
EJMECH 9836
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 17 July 2017
Revised Date: 28 September 2017
Accepted Date: 15 October 2017
Please cite this article as: E. Zhang, P.-Y. Bai, D.-Y. Cui, W.-C. Chu, Y.-G. Hua, Q. Liu, H.-Y. Yin,
Y.-J. Zhang, S. Qin, H.-M. Liu, Synthesis and bioactivities study of new antibacterial peptide mimics:
The dialkyl cationic amphiphiles, European Journal of Medicinal Chemistry (2017), doi: 10.1016/
j.ejmech.2017.10.044.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Graphical abstract

ACCEPTED MANUSCRIPT
Synthesis and bioactivities study of new antibacterial peptide mimics:
the dialkyl cationic amphiphiles
1
,2
1
1
1
1
En Zhang* , Peng-Yan Bai , De-Yun Cui , Wen-Chao Chu ,Yong-Gang Hua , Qin
1
1
1
1,2
1,2
Liu , Hai-Yang Yin , Yong-Jie Zhang , Shangshang Qin* , Hong-Min Liu*
1
School of Pharmaceutical Sciences; Institute of Drug Discovery and Development; Key
Laboratory of Advanced Pharmaceutical Technology, Ministry of Education of China; Zhengzhou
2
University, Zhengzhou 450001, PR China. Collaborative Innovation Center of New Drug
Research and Safety Evaluation, Henan Province, Zhengzhou 450001, PR China.
E-mail: zhangen@zzu.edu.cn; qinshangshang@126.com; liuhm@zzu.edu.cn
ABSTRACT
The emergence of infectious diseases caused by pathogenic bacteria is widespread.
Therefore, it is urgently required to enhance the development of novel antimicrobial
agents with high antibacterial activity and low cytotoxicity. A series of novel dialkyl
cationic amphiphiles bearing two identical length lipophilic alkyl chains and one
non-peptidic amide bond were synthesized and tested for antimicrobial activities
against both Gram-positive and Gram-negative bacteria. Particular compounds
synthesized showed excellent antibacterial activity toward drug-sensitive bacteria
such as S. aureus, E. faecalis, E. coli and S. enterica, and clinical isolates of
drug-resistant species such as methicillin-resistant S. aureus (MRSA), KPC-producing
and NDM-1-producing carbapenem-resistant Enterobacteriaceae (CRE). For example,
the MIC values of the best compound 4g ranged from 0.5 to 2 µg/mL against all these
strains. Moreover, these small molecules acted rapidly as bactericidal agents, and
functioned primarily by permeabilization and depolarization of bacterial membranes.
Importantly, these compounds were difficult to induce bacterial resistance and can
potentially combat drug-resistant bacteria. Thus, these compounds can be developed
into a new class of antibacterial peptide mimics against Gram-positive and
Gram-negative bacteria, including drug-resistant bacterial strains.
Keywords: Antimicrobial; dialkyl cationic amphiphiles; non-peptidic amide bonds;

ACCEPTED MANUSCRIPT
drug-resistant; permeabilization; depolarization
. Introduction
Antimicrobials are probably one of the most successful forms of chemotherapy
1
in the history of medicine. These compounds have saved many lives and contributed
significantly to the control of infectious diseases.[1] However, prevailing bacterial
resistance to antibiotics combined with stagnant antibiotic discovery has created an
urgent need for the development of new antimicrobial agents that exert novel
mechanisms of action.[2, 3]
Because of the widespread use and misuse of antimicrobial agents, the
multidrug-resistance of bacteria has become a severe threat to public health.[4] There
are many pathogenic bacteria that evade antibiotics, notably the “ESKAPE” bacteria:
E. faecium, S. aureus, K. pneumoniae, A. baumanii, P. aeruginosa and Enterobacter
species.[5] Since MRSA was first reported[6] in the 1950s, it has gradually become
the most commonly identified drug-resistant bacteria worldwide.[7] Quaternary
ammonium compounds (QACs) were widely used to eradicate bacteria, but the
presence of qac genes in MRSA led to the emergence of QACs resistance.[8-12] The
carbapenem-resistant strains are also very problematic. Such as New Delhi
metallo-β-lactamase 1 (NDM-1) producing isolates, which are highly resistant to
nearly all antibiotics, except colistin and tigecycline.[13] And KPC-2 carbapenemase
producing isolates were also exhibited resistance to all β-lactam agents and other
antimicrobials.[14] Although the number of bacteria showing antibiotic resistance
grows, between 1983 and 2012 the number of new systemic antibacterial agents
approved by the US Food and Drug Administration continued to decrease.[15] Thus,
there is an urgent need to develop novel antibacterial agents against such bacteria.
Antimicrobial peptides (AMPs) have received substantial attention in recent
years owing to their membrane targeting mechanism, broad-spectrum activities and
minimum cytotoxicity.[16-19] AMPs display a diverse range of primary peptide
sequences and secondary structures. Additionally, AMPs have two characteristics: a
large number of cationic residues and hydrophobic domains.[20] Several AMPs are
currently being validated in preclinical and clinical settings for the treatment of

ACCEPTED MANUSCRIPT
infections caused by antimicrobial-resistant bacteria.[21] However, AMPs have some
disadvantages, including in vivo toxicity, protease instability and high manufacturing
costs.[20] Therefore, a variety of structurally diverse synthetic membrane-active
compounds have also been developed as an alternative to antibacterial peptides, such
as peptidomimetics[22], oligomers[23], antimicrobial polymers (e.g., cationic
derivatives of polymethacrylate[24, 25], polyarylamide[26] and polystyrene[27]),
cationic amphiphiles[28, 29] and β-2,2-amino acid derivatives[30].
Several small amphiphilic molecules based on natural product skeletons of
xanthone[31, 32] and binaphthyl[33] with potent antibacterial and antibiofilm
activities have been developed.[34-39] Additionally, amphiphilic derivatives based on
approved antibiotics such as vancomycin[40-42] and kanamycin B[43],[44] were
reported to have potent antimicrobial activity. There are already many synthetic
antibacterial peptidomimetics currently undergoing clinical trials, such as the
membrane-disrupting compound LTX 109[45] (Fig. 1), the cationic steroid compound
CSA-13[46] (Fig. 1) and the novel peptide mimetic brilacidin[47] (Fig. 1). These
synthetic analogues, possessing excellent antimicrobial activity and high selectivity,
were relatively facile and inexpensive to prepare in large quantities. The diversity of
structures of these molecules requires further exploration.[26]
Fig. 1. Structures of some representative AMPs.
During the course of our recent investigations for new structures with good
biological activity,[48, 49] we disclosed several heterocyclic derivatives with good
antimicrobial activity.[50-52] Inspired by these exciting results, we further showed

ACCEPTED MANUSCRIPT
through preliminary results the broad antibacterial activity of small amphiphilic
molecules[29, 39] against several important drug-resistant strains.
Recently, a series of small molecular cationic compounds with membrane-active
and selective broad-spectrum activity were designed and synthesized. The preparation
of these new antibacterial amphiphiles was through an effective and reliable synthetic
method. The final compounds were obtained by two or three step processes.[35]
Moreover, these compounds possessed several positive charges, had two lipophilic
alkyl chains and one nonpeptidic amide bond, and displayed good solubility in water.
Herein, we reported the activity of these compounds against drug-sensitive bacteria,
including S. aureus, E. coli, E. faecalis and S. enterica, and drug-resistant bacteria
such as methicillin-resistant S. aureus (without of qac genes), NDM-1-producing
Enterobacteriaceae (NDM) and carbapenemase-producing K. pneumoniae (KPC).
The bactericidal kinetics and the propensity of bacteria to develop resistance against
these compounds were studied. The mechanism of action of the compounds to disrupt
bacterial biofilms was investigated. Additionally, stability and activity of these
molecules in complex mammalian fluids such as plasma, serum and whole blood were
evaluated to understand the optimal conditions that enhanced the potency of these
small molecules. The toxicity of the compounds was studied using HeLa cells and
mammalian red blood cells. Interestingly, using fluorescence and electron scanning
spectroscopy, these amphiphilic compounds did not lead to cell death at the minimal
inhibitory concentration.
Scheme 1. Synthesis of symmetric dialkyl amine intermediates (1a–1i).

ACCEPTED MANUSCRIPT
(
a) DIPEA, HBTU in DMF/CHCl (5:2), RT, 24 h; (b) CH COCl/CH OH, 0 °C–RT,
3
3
3
2
4 h.
Scheme 2. Synthesis of dialkyl cationic amphiphiles (4a–4u).
2
2
. Results and discussion
.1. Synthesis and characterization
The molecules 4a–4u were synthesized only in three steps using amines as
starting materials (Schemes 1 and 2). To assess the importance of the amphiphilic
balance, the length of the lipophilic alkyl chains was varied from two to 11. To further
fine-tune the structure-activity relationship of the compounds, hydrophilic moieties
(
amino acids) and the configuration of the amino acids were also varied on these lipid
frameworks. To synthesize these compounds, the various intermediates 2a, 2e–2i,[39]
b,[53] 2c and 2d[54] were prepared by known methods. The general syntheses of
N,N-dialkylamines (1e 1g and 1i; Scheme 1) were accomplished by a
2
,
conventional procedure[55] and other N,N-dialkylamines were commercial
available. Key amide coupling compounds (3a–3u; Scheme 2) were synthesized

ACCEPTED MANUSCRIPT
using N-Boc-protected amino acids and 1a–1i. Finally, target compounds (4a–4u;
Scheme 2) were obtained individually from 3a–3u with acetyl chloride in methanol.
The products were obtained without further chromatography purification in high yield.
1
13
All final compounds were characterized by H NMR, C NMR and HRMS. Finally,
1 dialkyl cationic small molecules were synthesized using different lipophilic
secondary amines with the same alkyl chain and nine different amino acids.
2
2
.2. Antibacterial activity
The antibacterial efficacy of these compounds was determined in suitable culture
medium and expressed as the minimum inhibitory concentration (MIC) (Table 1). The
glycopeptide antibiotic vancomycin and the β-lactam antibiotic meropenem were also
used to compare the MIC results with our compounds. These compounds displayed
preferable antibacterial potency against a wide spectrum of drug-sensitive bacteria
such as S. aureus, E. coli, E. faecalis and S. enterica, and drug-resistant bacteria such
as methicillin-resistant S. aureus (MRSA), NDM-1-producing Enterobacteriaceae
and carbapenemase-producing K. pneumoniae.
In general, the compounds 4e–4m, 4o, 4q and 4u showed good activities that
were comparable to vancomycin against all the bacteria tested, and these cationic
small molecules were found to be more active toward Gram-positive bacteria than
Gram-negative bacteria. The range of MIC values for the small molecules (4e–4m, 4o,
4
q and 4u) were 0.5–8 µg/mL against Gram-positive drug-sensitive and drug-resistant
bacteria, whereas the range of MIC values was 1–32 µg/mL for Gram-negative
bacteria. The MICs of compounds 4e–4m, 4o, 4q and 4u were comparable to MSI-78
(an AMP currently undergoing phase III clinical trials as a topical antibiotic).[56, 57]
The amphiphilic balance of the small molecules was evaluated by ClogP and was
found to correlate very well with their antibacterial activity. The ClogP values of
these compounds ranged from 0.96 to 9.48 (Table 1). The antibacterial activity was
good when the ClogP values ranged between 5 and 8. For example, the span of MIC
values for the compounds 4m and 4o were 2–32 µg/mL and their ClogP value was
6
.62. In contrast, the range of MIC values for molecules 4n and 4p were 8–128
µg/mL and even >128 µg/mL, and their ClogP value was 8.73. Additionally, the

ACCEPTED MANUSCRIPT
configuration of the amino acid showed negligible influence to the biological activity.
4
m and 4o with L-serine had similar activity compared with that of 4n and 4p with
D-serine residues, which is consistent with literature.[37]
Table 1. MIC and hemolytic activity (HC ) of small molecules.
5
0
MIC µg/mL
Drug-sensitive bacteria
e
HC50
ClogP
a
Drug-resistant bacteria
(µg/mL)
E.
faecalis
E. coli
S.
MRSA
KPC
compd
S. aureus
NDM
enterica
b
b
b
4
4
a
>128
>128
128
128
64
8
>128
>128
128
64
8
>128
128
128
64
-
-
-
0.96
2.07
3.13
4.19
5.25
6.31
7.36
8.42
9.48
4.89
5.95
7.01
6.62
8.73
6.62
8.73
7.57
8.02
8.02
8.98
6.92
>1000
743
b
b
b
b
>128
128
64
8
-
-
-
b
b
b
4
c
-
-
-
576
b
b
b
4
d
-
-
-
496
4
e
16
4
0.5
0.5
1
16
1
16
4
434
4
f
2
2
2
4
225
4
4
g
1
1
2
2
2
2
133
h
4
2
2
2
2
4
98
4
i
4
2
16
2
8
4
8
32
4
89
4
j
1
2
8
0.5
2
4
124
4
k
1
2
2
4
4
4
119
4
l
2
2
8
4
2
4
4
87
4
m
4
2
8
8
2
32
16
>1000
>1000
367
b
b
4
n
o
p
q
8
8
>128
8
>128
8
16
4
-
-
4
4
4
32
16
b
b
4
4
16
4
32
8
>128
4
>128
32
32
4
-
-
>1000
306
16
64
b
b
4
r
s
2
2
64
>128
>128
16
>128
>128
>128
64
4
-
-
>1000
293
b
b
4
4
4
4
-
-
b
b
4
t
8
128
8
32
8
-
-
>1000
268
4
u
4
64
64
c
b
b
b
b
b
b
b
b
VAN
2
-
-
-
-
-
-
-
-
d
b
b
b
b
b
b
b
b
MEM
-
-
<0.125
-
-
-
-
-
-
f
f
f
b
f
b
b
b
g
MSI-78
8-16
64
16-32
-
16-32
-
-
-
120
a
MRSA (methicillin-resistant S. aureus), KPC (carbapenemase-producing K. pneumoniae), NDM-1
b
c
d
e
(
(
NDM-1-producing Enterobacteriaceae), f not determined. VAN (vancomycin), MEM (meropenem), ClogP
g
calculated with ChemBioDraw software), literature values[57], literature values[58].

ACCEPTED MANUSCRIPT
Fig. 2. MICs of cationic small molecules: (a) with varying alkyl chain lengths (4a–4i);
(
b) with different amino acids (4f, 4j, 4m, 4o and 4q–4u). The black star represents
MIC >128 µg/mL.
The length of the alkyl chain of compounds 4a–i (Fig. 2a) was extremely
important to their antibacterial activity. The MICs were good when the lipophilic
chain was C H , C H and C H . This is consistent with reported results.[37, 39]
7
15
8
17
9
19
Therefore, the antibacterial activity of these compounds was affected strongly by the
length of the alkyl chain.
The kinds of amino acids attached to the compounds had an important influence
on antibacterial activity. For example, compounds (4f, 4j, 4m, 4o, 4q–4u; Fig. 2b)
with the same lipophilic moieties and different hydrophilic amino acids displayed
significant differences in their antibacterial activity. Compounds 4r, 4s and 4t showed
no activity toward Gram-negative bacteria, but molecule 4r had good MICs against
Gram-positive bacteria (2 µg/mL against S. aureus, 2 µg/mL against E. faecalis),
which was comparable to vancomycin activity (2 µg/mL against S. aureus) against
Gram-positive bacteria. Compound 4s also had similar selectivity, and this is the first
report on small cationic antimicrobial peptides with an L-leucine residue (4r). The
antimicrobial activity of compounds 4r, 4s and 4t was reduced when the amino acid
residues were L-leucine, L-isoleucine, or L-phenylalanine. The structure-activity
relationship studies of this series of dialkylamine cationic small molecules revealed
that an amphiphilic structure was critical for broad-spectrum activity, especially
against Gram-negative bacteria.
Table 2. MIC of small molecules against drug-resistant MRSA.
MRSA
compound

ACCEPTED MANUSCRIPT
M-1 M-2 M-3
M-4
16
1
M-5
8
M-6
4
M-7
4
M-8 M-9 M-10
4e
8
1
16
1
16
1
8
1
4
1
4
0.5
0.5
1
4f
1
0.5
1
0.5
1
4g
1
1
1
2
1
1
1
4h
1
1
2
2
2
1
1
1
1
4i
2
1
2
2
4
4
4
4
4
4
4
j
1
1
0.5
1
1
1
0.5
2
0.5
2
1
1
0.5
2
4k
2
1
1
2
2
2
4l
2
1
2
2
2
2
2
2
2
2
4m
4
4
4
4
4
4
2
4
4
2
4
n
o
p
q
16
4
4
4
4
16
4
8
8
16
4
16
4
16
4
4
4
4
4
4
4
4
4
32
4
32
4
32
4
32
4
32
4
32
4
32
4
32
4
32
4
32
4
4r
4
2
2
2
4
4
4
4
4
4
4s
4
4
4
4
4
4
4
4
4
4
4
t
32
8
16
4
16
4
16
4
32
8
32
8
16
8
32
8
32
8
32
8
4u
The antibacterial activity (Table 2) of selected small molecules against
drug-resistant MRSA (10 clinical isolated strains) was further investigated. All
selected molecules had good MIC values. Among the 17 compounds in this
experiment, 4j was most effective. In addition, this series of small cationic
amphiphiles were extremely active against MRSA with MICs of ~0.5–4 µg/mL. The
MIC of 4j (0.5–1 µg/mL) was comparable to vancomycin (0.9 µg/mL)[39] and better
than MSI-78 (16–32 µg/mL)[57] (Table 1).
Table 3. MIC of small molecules against drug-resistant bacteria KPC.
KPC
Compound
K-1 K-2
K-3
16
2
K-4
16
2
K-5
16
2
K-6
16
2
K-7
16
2
K-8
16
1
K-9 K-10
4e
16
2
16
2
16
2
16
2
4f
4g
2
2
2
2
2
2
2
2
2
2
4h
4
4
4
4
4
4
4
2
4
4
4
i
32
8
16
4
16
4
8
32
8
32
8
32
8
8
32
8
32
4
4j
4
4
4k
4
4
4
2
4
4
4
4
4
4
4l
4
4
4
4
4
4
4
4
4
4
4
m
32
32
16
64
32
32
16
64
32
32
16
64
32
32
32
64
32
32
32
64
32
32
32
64
32
32
32
64
32
32
16
64
32
32
16
64
32
32
16
64
4
o
q
u
4
4

ACCEPTED MANUSCRIPT
The antibacterial activity (Table 3) of small molecules against KPC (10 clinical
isolated strains) was tested. The MICs of these compounds ranged from 1–64 µg/mL.
The MIC of the most effective compounds 4f–4h and 4j–4l were 2–8 µg/mL. Most
importantly, compounds 4f–4h and 4j–4l were even better than MSI-78 (8–16
µg/mL).
Table 4. MIC of small molecules against drug-resistant bacteria NDM-1.
NDM-1
Compound
N-1 N-2
N-3
32
8
N-4
32
8
N-5
16
4
N-6
16
4
N-7
32
8
N-8
32
8
N-9 N-10
4
e
32
8
32
8
16
8
16
8
4
f
4
4
g
2
2
2
2
2
2
2
2
4
4
h
4
4
4
4
4
4
4
4
4
4
4
i
32
8
32
8
32
8
32
8
32
4
32
4
16
4
16
4
32
8
32
8
4
j
4
k
4
4
4
4
4
4
2
2
4
4
4
l
4
4
4
4
4
4
4
4
8
8
4
m
32
32
64
32
16
64
16
16
64
64
16
16
64
64
16
16
64
64
16
16
64
>64
32
32
64
>64
32
32
64
>64
32
16
64
>64
32
16
64
>64
4
o
q
u
4
4
>64 >64
Moreover, the antibacterial activity (Table 4) of small molecules against NDM-1
(10 strains) was investigated. The MICs ranged from 2 to 64 µg/mL and even >64
µg/mL. The MIC of the most effective compounds 4f–4h and 4j–4l were 2–8 µg/mL.
From the MIC values (Tables 1–4), compounds 4f–4h and 4j–4l not only had
potent activity against both drug-sensitive bacteria and drug-resistant bacteria, but
also had comparative activity against both Gram-positive and Gram-negative bacteria.
Furthermore, the antibacterial activities of compounds 4f–4h and 4j–4l were better
than the corresponding activity of MSI-78. Compounds 4f–4h and 4j–4l showed
broad-spectrum antibacterial activity, whereas other molecules had better activities
against Gram-positive bacteria than against Gram-negative bacteria. The structures of
molecules 4f–4h and 4j–4l have two alkyl chains (-C H , -C H , -C H ) with an
7
15
8
17
9
19
L-lysine or L-arginine residue. The ClogP values of amphiphiles 4f–4h and 4j–4l
ranged between 5 and 8.4. The optimum amphiphilicity may be an essential element
to achieve the maximum antibacterial activity of these amphiphilic small molecules.

ACCEPTED MANUSCRIPT
Noteworthy, compounds 4r and 4s had selective antibacterial activity against
Gram-positive bacteria.
Fig. 3. 50% hemolysis concentration of cationic small molecules: (a) with varying
lipophilic alkyl chain (compounds 4a–4i); (b) with varying amino acids (compounds
4
f, 4j, 4m, 4o and 4q–4u). The black star represents >1000 µg/mL.
.3. Hemolytic activity
2
The hemolytic activity of the cationic small molecules 4a–4u was represented as
their HC values. The ability of compounds to lyse red blood cells (RBCs) was the
5
0
toxicity evaluation of compounds toward mammalian cells. HC values of the small
50
molecules ranged between 87 and >1000 µg/mL (Table 1). In general, HC values of
5
0
the cationic small molecules 4a–4i with different lipophilic alkyl chains were found to
increase as the alkyl chain length increased (Fig. 3a). For example, the HC values of
5
0
compounds 4a–4i were >1000, 743, 576, 496, 434, 225, 133, 98 and 89 µg/mL,
respectively. To our delight, compounds 4a–4i showed minimal toxicity toward red
blood cells. On the other hand, the hemolytic activity of compounds 4f, 4j, 4m, 4o
and 4q–4u with different amino acids and the same alkyl chain length were found to
vary considerably (Fig. 3b). For example, the HC values of compounds 4m, 4r and
5
0
4
t were >1000 µg/mL, whereas compounds 4f, 4j, 4o, 4q, 4r, 4s and 4u ranged
between 124 and 367 µg/mL. However, compound 4g, one of the most potent
molecules, had an HC value of 133 µg/mL, giving good selectivity (S = HC /MIC)
5
0
50
against S. aureus (133) and E. coli (66.5) respectively (Table 1). Compound 4r had
excellent selective antibacterial activity (MIC = 2 µg/mL against both S. aureus and E.

ACCEPTED MANUSCRIPT
coli) against Gram-positive bacteria and low hemolytic activity (HC >1000 µg/mL).
5
0
The hemolytic toxicity study revealed that these compounds had minimal toxicity
toward mammalian red blood cells and the above results indicate that the activity of
these compounds was selective toward bacteria.
Fig. 4. Plasma stability and antibacterial activity in complex mammalian fluids of
small molecules: (a) antibacterial efficacy of compounds 4f, 4g, 4h, 4j, 4k and 4l
against S. aureus after preincubating in 50% plasma for different periods of time (0, 3
and 6 h); (b) minimum bactericidal concentrations (MBCs) of 4f, 4g, 4h, 4j, 4k and 4l
in 50% serum, 50% plasma and 50% blood against MRSA.
2
.4. Plasma stability
Protease degradation is a major disadvantage of natural antibacterial peptides,
which results in reduced antibacterial activity of these peptides in mammalian
[59]
fluids.
molecules under plasma conditions, antibacterial efficacy of compounds 4f, 4g, 4h, 4j,
k and 4l against S. aureus after preincubating in 50% plasma for different periods of
To determine the stability of the dialkylamine cationic amphiphilic
4
time (0, 3 and 6 h) was evaluated. The MBC values of compounds 4f, 4g, 4h, 4j, 4k
and 4l increased from 4 or 8 µg/mL (100% media) to 16 µg/mL (50% plasma) after 0,
3
and 6 h treatment (Fig. 4a). The above results indicate that compounds 4f, 4g, 4h, 4j,
4
k and 4l lost only a small amount of antibacterial efficacy upon pretreatment in
plasma.
2
.5. Antibacterial activity in complex mammalian fluids
The loss of antibacterial activity of the compounds in the presence of complex

ACCEPTED MANUSCRIPT
mammalian fluids was another serious concern. Antibacterial activity of compounds
4
f, 4g, 4h, 4j, 4k and 4l was investigated by determining the minimum bactericidal
concentration (MBC) in 50% serum, 50% plasma and 50% blood supplemented with
0% MHB against MRSA. The compounds 4f, 4g, 4h, 4j, 4k and 4l were found to be
5
active in 50% serum and 50% plasma, but were not active in 50% blood. The MBC
values of compounds 4f, 4g, 4h, 4j, 4k and 4l increased 1-fold or 2-fold in 50% serum
and 2-fold or 4-fold in 50% plasma, whereas the MBC values increased 8-fold or
1
6-fold in 50% blood (Fig. 4b). The 2–4-fold increase of MBC values could be due to
negatively charged proteins and macromolecules in human serum, plasma, or blood
that tightly bind to the cationic molecules, thereby inhibiting their activity toward
bacterial membranes. The above results indicate that compounds 4f, 4g, 4h, 4j, 4k and
4
l were active in complex mammalian fluids like serum and plasma, but these
compounds were not able to retain their antibacterial activity in blood.
Fig. 5. Time-dependent killing of pathogens by compound 4g. (a), (c) S. aureus were
grown to early (a) and late (c) exponential phase, and challenged with compound 4g
(at 6× MIC and 8× MIC) and Van (vancomycin); (b), (d) E. coli were grown to early

ACCEPTED MANUSCRIPT
(b) and late (d) exponential phase, and challenged with compound 4g (at 6× MIC and
8
× MIC) and Mox (moxalactam). The control was treatment with sterile water. Data
are representative of three independent experiments.
2
.6. Bactericidal time-kill kinetics
The time-kill kinetics of compound 4g was performed to determine the rate of
bactericidal action. This investigation was carried out against S. aureus and E. coli. S.
aureus and E. coli were grown to early and late exponential phase and challenged
with compound 4g (at 6× MIC and 8× MIC), vancomycin (S. aureus) and
moxalactam (E. coli). Compound 4g had excellent bactericidal activity against S.
aureus, showing superior activity to vancomycin in killing early and late exponential
phase populations (Fig. 5a, c). Moreover, 4g killed early exponential phase S. aureus
at 0.5 h at 8× MIC and 120 min at 6× MIC, and 4g killed late exponential phase S.
aureus at 6 h at 8× MIC and 6× MIC, while vancomycin did not kill S. aureus at 6 h.
Compound 4g displayed rapid bactericidal activity against E. coli, and was superior to
moxalactam in killing early and late exponential phase bacteria (Fig. 5b, d). 4g killed
early and late exponential phase E. coli at 0.5 h and at 1 h, respectively, while
moxalactam did not kill early or late exponential phase E. coli at 6 h. Clearly,
compound 4g had excellent antibacterial activity against S. aureus and E. coli (Fig. 6),
and after 6 hours treatment the control group samples were cloudy, but the groups of
4
g (at 6× MIC and 8× MIC) were both clear. Thus, the results suggest that the dialkyl
cationic amphiphile 4g killed both Gram-positive and Gram-negative bacteria rapidly.
Fig. 6. Time-kill kinetics of compound 4g. (a), S. aureus were grown to early
exponential phase and challenged with compound 4g (at 6× MIC and 8× MIC) and
vancomycin; (b), E. coli were grown to early exponential phase and challenged with

ACCEPTED MANUSCRIPT
compound 4g (at 6× MIC and 8× MIC) and moxalactam. The control was sterile
water. Pictures taken after treatment with compounds for 6 h.
Fig. 7. Antibiofilm activity of small molecule 4g. (a) Cell viability in non-treated and
treated biofilms of S. aureus grown on cover slips for 24 h and after treating the
biofilms with 4g at different concentrations. (b) Cell viability in non-treated and
treated biofilms of E. coli grown on cover slips for 72 h and after treating the biofilms
with 4g at different concentrations. The black star represents <50 CFU/mL.
2
.7. Biofilm disruption activity
To evaluate the efficiency of this class of compounds to eradicate preformed
biofilms, one of the best compound 4g was used against established S. aureus and E.
coli biofilms. Mature S. aureus biofilms in a 96-well plate (grown for 24 h) with an
initial count of 21.9 log CFU/mL per well of bacteria were treated with 4g at seven
10
different concentrations. 4g was found to reduce cell viabilities in the biofilms (19.9,
8.6, 11.9, 11.5, 9.9, 7.9 and 7.6 log CFU per well at 2, 4, 8, 16, 32, 64 and 128
1
1
0
µg/mL, respectively), whereas the cell viability in the non-treated biofilm increased to
4.9 log CFU per well (Fig. 7a). 4g was also able to reduce the cell viabilities in
2
1
0
mature E. coli biofilms (developed for 72 h) from an initial count of 25.9 log CFU
1
0
per well to 24.7, 23.9, 8.4, 4.1, 2.1 and 0.3 log CFU per well at 2, 4, 8, 16, 32, 64
1
0
and 128 µg/mL, respectively, whereas the cell viability in the non-treated biofilm
increased to 28.5 log10 CFU per well (Fig. 7b). Biofilm disruption activity of

ACCEPTED MANUSCRIPT
compound 4g was visually shown by crystal violet staining (Fig. 8a, b). Moreover, the
MBIC values of 4g were 18.0 µg/mL against S. aureus and 53.9 µg/mL againist E.
8
0
coli (Table 5). And the MBEC values of 4g were 64 µg/mL and 128µg/mL against S.
aureus and E. coli, respectively. That were 64-fold of MIC values. The above results
and images indicate that compound 4g can inhibit bacterial biofilm formation and
eradicate established biofilms at 64 and 128 µg/mL.
Fig. 8. Antibiofilm activity of small molecule 4g. (a) Images of the small molecule
treated and non-treated biofilms of S. aureus after staining with crystal violet. (b)
Images of the small molecule treated and non-treated biofilms of E. coli after staining
with crystal violet.
Table 5. MBIC and MBEC of compound 4g.
MBIC / MBIC (µg/mL)
MBEC (µg/mL)
8
0
90
Compound
S. aureus
E. coli
53.9/73.6
S. aureus
64
E. coli
128
4
g
18.0/26.9
2
.8. Mechanism of action
The detailed molecular mechanism of action was investigated to confirm whether
these dialkyl cationic amphiphilic small molecules acted by disrupting the integrity of
the bacterial cell membrane. Gram-positive S. aureus and Gram-negative E. coli were
used with all compounds to assess the structure-activity relationship of this series of
compounds. The mechanism of action can be confirmed by the following three
experiments (membrane depolarization, and inner membrane and outer membrane
permeabilization, Fig. 9).

ACCEPTED MANUSCRIPT
Fig. 9. Mechanism of antibacterial action of the cationic small molecules (4a–4u).
Cytoplasmic membrane depolarization of S. aureus (a) and E. coli (b). Inner
membrane permeabilization of S. aureus (c) and E. coli (d). (e) Outer membrane
permeabilization of E. coli in the presence of amphiphilic small molecules at 10
µg/mL.
2
.8.1. Cytoplasmic membrane depolarization
The dialkyl amphiphilic small molecules were found to dissipate the membrane
potential of both Gram-positive and Gram-negative bacteria, as monitored by the

ACCEPTED MANUSCRIPT
membrane-potential fluorescence sensitive dye diSC35. Furthermore, different
compounds showed different levels of dissipation of the membrane potential.
Compounds 4g, 4k, 4l, 4h and 4o (final concentration was 20 µg/mL) showed
maximum membrane depolarization, whereas compounds 4a–4e, 4n and 4p exhibited
less membrane depolarization against both S. aureus and E. coli (Fig. 9a, b). The
results further indicate that the ability of the compounds 4a–4u to dissipate the
membrane potential are well correlated with their MIC values.
2
.8.2. Inner membrane permeabilization
Bacterial cytoplasmic membrane permeabilization was studied using the
fluorescent probe propidium iodide (PI). PI enters bacteria only through compromised
membranes and fluoresces upon binding to cellular DNA. After treatment with the
small molecules 4a–4u (final concentration was 20 µg/mL), an enhancement in the
fluorescence intensity was observed in S. aureus and E. coli (Fig. 9c, d). Thus, the
results show that compounds 4a–4u permeabilize the membrane of both
Gram-positive and Gram-negative bacteria.
2
.8.3. Outer membrane permeabilization
Outer membrane permeabilization was studied using the hydrophobic dye
N-phenyl naphthylamine (NPN). NPN was generally excluded from the outer
membrane of Gram-negative bacteria. When the outer membrane was damaged, NPN
accessed the impaired outer membrane, exhibiting an increase in fluorescence
intensity. After treatment with the small molecules 4a–4u (final concentration was 20
µg/mL), an enhancement in the fluorescence intensity was observed in E. coli;
compounds were able to permeabilize the outer membrane (Fig. 9e). The above
results indicate that compounds 4a–4u can permeabilize the membrane of
Gram-negative bacteria.

ACCEPTED MANUSCRIPT
Fig. 10. Electron scanning microscopy images of HeLa cells following treatment with
small molecules 4g and 4k for 24 h. (a) Cells treated with 4g (1× MIC); (b) cells
treated with 4k (2× MIC); and (c) cells treated with 0.1% Triton-X (positive control)
and (d) non-treated cells (negative control). Scale bar is 10 µm.

ACCEPTED MANUSCRIPT
Fig. 11. Fluorescence microscopy images of HeLa cells after treatment with small
molecules 4g and 4k for 24 h and staining with calcein AM and propidium iodide (PI).
(a-c) Non-treated cells (negative control); (d-f) cells treated with 4g (1× MIC); (g-i)
cells treated with 4k (2× MIC); and (j-l) cells treated with 0.1% Triton-X (positive
control). Scale bar is 200 nm.
2
.9. Fluorescence and electron scanning microscopy
Cells were seeded into the wells of a 12-well plate and then treated with
compounds (4g and 4k) at various concentrations (1, 2 and 4 µg/mL). The treated cell
lines were imaged by optical microscopy to visualize the morphology (Fig. 10). The
treated cells were found to have normal morphology at 1× MIC and 2× MIC, (Fig.
1
0a, b), and they were similar to the untreated cell lines (Fig. 10d). In contrast, cells
treated with Triton-X were found to have a completely damaged shape (Fig. 10c).
Fluorescent microscopy studies using the live/dead staining method showed that cells
treated with compounds 4g and 4k showed green fluorescence even at 1× MIC, 2×
MIC (Fig. 11d-f, g-i) and were similar to the untreated cells (Fig. 11a-c). In contrast,
cells treated with Triton-X were found to have completely red fluorescence (Fig.
1
1j-l). The above results indicate that these small molecules were indeed nontoxic
toward mammalian cells at their MIC values.
Fig. 12. Propensity of development of bacterial resistance against compound 4g. (a)
For S. aureus where antibiotic norfloxacin was used as the control; (b) for E. coli
where lipopeptide colistin was used as the control.
2
.10. Propensity to induce bacterial resistance

ACCEPTED MANUSCRIPT
It is important to evaluate the potential emergence of bacterial resistance against
these small molecules. We took one of the most active compounds (i.e., 4g) as a
representative to evaluate the ability of these compounds to suppress the development
of resistance against both Gram-positive S. aureus and Gram-negative E. coli. Two
control antibiotics norfloxacin and colistin were chosen for S. aureus and for E. coli,
respectively. In the case of norfloxacin and colistin, the initial MIC values were
determined against respective bacteria. After the initial MIC experiment, serial
passaging was initiated by transferring the growing bacterial suspension at sub-MIC of
the compound/antibiotics (at MIC/2), and was subjected to another MIC assay and the
process was repeated for 20 passages. The cationic compound 4g showed no change
in the MIC against both S. aureus and E. coli even after 20 passages, whereas
2
56-fold and 32-fold increases in MIC were observed for norfloxacin and colistin,
respectively (Fig. 12a, b). The above results indicate that these small molecules are
difficult to induce bacterial resistance and they represent potential compounds to
combat drug-resistant bacteria. In contrast, resistance to the two positive control
antibiotics norfloxacin and colistin by S. aureus and E. coli, respectively, was easily
developed.
3
. Conclusions
Twenty-one small molecules using different secondary amines with the same
alkyl chain and nine different amino acids were synthesized. These compounds
displayed potent broad-spectrum and membrane-active antibacterial activity against
various drug-sensitive and drug-resistant bacteria (both Gram-positive and
Gram-negative). Variation of the amino acids and amphiphilic nature of the dialkyl
cationic amphiphiles strongly affected their antibacterial activity as well as hemolytic
activity. The structure-activity relationship studies of this series of dialkyl cationic
small molecules revealed that an amphiphilic structure was critical for broad-spectrum
activity, especially against Gram-negative bacteria. Additionally, the small molecules
were found to be stable under plasma conditions and showed excellent activity in
various complex mammalian fluids such as serum and plasma, but these compounds
were not able to retain antibacterial activity when incubated in blood. Moreover, these

ACCEPTED MANUSCRIPT
small molecules acted on bacteria rapidly and did not allow bacteria to develop
resistance. Importantly, various spectroscopic and microscopic studies revealed that
depolarization and disruption of the bacterial cell membrane are the primary
mechanisms describing the bactericidal action of the compounds. Furthermore, the
optimized molecule was found to be nontoxic toward mammalian red blood cells and
showed negligible toxicity against mammalian cells. Therefore, these molecules hold
great promise to combat drug-resistance pathogens and have the ability not only to
inhibit bacterial biofilm formation but also to eradicate established biofilms. Thus,
these compounds can be potentially used to tackle infections caused by various
bacteria.
4
. Experimental section
General: Reagents and solvents were purchased from commercial sources and were
1
13
used without further purification. H NMR and C NMR spectra were recorded on a
Bruker 400 MHz and 101 MHz spectrometer respectively, and TMS as internal
standard reference. Coupling constants (J) are given in hertz (Hz). High resolution mass
spectra (HRMS) were recorded on a Waters Micromass Q-T of Micromass
spectrometer. The compounds were purified by reverse phase HPLC (Waters
Symmetry C-18 4.5×250mm, 5µm) using sodium dihydrogen phosphate buffer
(PH=3 or 7.5) /acetonitrile (0−100%) as mobile phase to more than 95% purity.
Analytical thin layer chromatography (TLC) was performed on glass plates pre-coated
with silica gel (5–40um, Qingdao Marine Chemical Factory (China)) to monitor the
reactions. Visualization was accomplished using UV light and 5% phosphomolybdic
acid ethanol solution. Column chromatography was performed on silica gel. For
optical density and fluorescence measurements, Tecan Infinite Pro series M200
Microplate Reader was used. The biological experiments were performed with a 1300
series A Biological Safety Cabinet. TDL-5M Desktop Low-speed Refrigerated
2
Centrifuge was used in antibacterial studies. All the solvents were of reagent grade.
Dimethylformamide, Dichloromethane, 1,2-Dichloroethane, Tetrahydrofuran,
Diethyl ether, Chloroform, Methanol, NaOH, K CO and other solvents were
2
3,

ACCEPTED MANUSCRIPT
supplied by chemical reagent company (China). Di-tert-butylcarbonate,
Diisopropylethylamine, HBTU, L-lysine, L-arginine, L-serine, D-serine, L-alanine,
L-leucine, L-isoleucine, L-phenylalanine and L-threonine were purchased from
Aladdin (China). N-n-heptylamine, n-nonylamine, n-undecylamine, Di-n-ethylamine,
di-n-butylamine, di-mylamine, di-n-hexylamine, di-n-octylamine and di-n-decylamine
were purchased from Chembee reagent company (China). All the chemicals were used
as supplied. Drug-sensitive bacterial strains, E.coli (ATCC 25922), S.aureus (ATCC
2
9213), E.faecalis (ATCC 29212) and S.enterica (ATCC 8387). Drug-resistant
bacteria, Methicillin-resistant S. aureus (MRSA), Carbapenemase-producing K.
pneumonia (KPC) and NDM-1-producing Enterobacteriaceae (NDM-1) were clinical
isolated strains. Sheep RBCs were used for hemolytic assay.
General procedure for the synthesis o f N, N-di-n-alkylamines ( 1e, 1g, 1i):
Compounds (1a-1d, 1f, 1h) were from commercial resources. The general synthesis
of other di-n-alkylamines (1e, 1g, 1i) are accomplished by conventional chemical
synthesis procedure as described below. Mixture of n-ethyl-amine (1 g, 1equiv),
n-ethyl bromide (1equiv) and anhydrous potassium carbonate (1equiv) in 7.5
mL dimethyl sulfoxide was kept under stirring for 12 h at 80 ºC.[55] And the
reaction was monitored by TLC, R = 0.2~0.3 (PE: EA: TEA = 5: 1: 0.05, v/v/v
)
f
and the visualization was accomplished using UV light and 5% phosphomolybdic acid
ethanol solution. PE represents petroleum ether, EA represents ethyl acetate and
TEA is triethylamine. The reaction mixture was taken in 25 mL chloroform,
washed with water (3 × 25 mL), dried over anhydrous magnesium sulfate and
filtered. Chloroform was removed from the filtrate on a rotary evaporator and the
resulting residue was purified by Silica gel column chromatography (PE: EA: TEA
=
10: 1: 0.05, v/v/v), then afforded the intermediate secondary amine.
1
N, N-di-n-heptylamine (1e): Colourless liquid, yield: 36.4%. H NMR (400 MHz,
CDCl3)
δ
2.62 – 2.54 (m, 4H), 1.55 – 1.41 (m, 4H), 1.29 (d,
J
= 3.3 Hz, 16H), 0.88 (t,
1
3
J
= 6.8 Hz, 6H). C NMR (101 MHz, CDCl )
δ
50.18, 31.83, 30.23, 29.27, 27.40,
3
2
2.62, 14.07.

ACCEPTED MANUSCRIPT
N, N-di-n-nonylamine (1g): Light yellow liquid, yield: 32.6%. H NMR (400
MHz, CDCl3) 2.61 – 2.54 (m, 4H), 2.21 (s, 1H), 1.47 (dd, = 13.7, 6.8 Hz, 4H), 1.28
d, = 6.0 Hz, 24H), 0.88 (t, = 6.8 Hz, 6H). C NMR (101 MHz, CDCl ) 50.11,
1.86, 30.16, 29.57, 29.55, 29.26, 27.40, 22.63, 14.02.
N, N-di-n-undecylamine (1i): White solid, yield: 32.5%. H NMR (400 MHz,
CDCl3) 2.61 – 2.56 (m, 4H), 1.48 (dd, = 13.9, 7.0 Hz, 4H), 1.27 (d, = 8.2 Hz,
= 6.8 Hz, 6H). C NMR (101 MHz, CDCl ) 50.11, 31.92, 30.10,
1
δ
J
1
3
(
J
J
δ
3
3
1
δ
J
J
1
3
3
2
2H), 0.88 (t,
9.63, 29.62, 29.60, 29.35, 27.43, 22.69, 14.11.
General procedure for the synthesis of intermediate molecules (2a, 2e-2i):
J
δ
3
L-lysine, L-alanine, L-leucine, L-isoleucine, L-phenylalanine, L-threonine (5 g, 1
equiv) was dissolved in H O (100 mL), and to it NaOH (3 equiv) was added and
2
stirred. To this, di-tert-butyl dicarbonate (Boc O) (2.4 equiv) in 50 mL of
2
tetrahydrofuran (THF) was added at 0 °C.[39] Then the reaction mixture was stirred
at room temperature for 24 h. At the end of the reaction, THF was removed under
reduced pressure and the aqueous layer was washed with diethyl ether to remove
organic impurities. Then the aqueous layer was acidified to pH 4~5 using 1M H SO
2
4
aqueous solution. The aqueous layer was then extracted with dichloromethane
DCM). The organic layer was then washed with brine and dried over anhydrous
Na SO . The organic layer was removed under reduced pressure to obtain the
(
2
4
compound.
Boc-Lys(Boc)-OH (2a): Light yellow solid, yield: 64.3%. H NMR (400 MHz, CDCl )
1
3
δ
5.30 (s, 1H), 4.73 (s, 1H), 4.21 (d,
J
= 73.0 Hz, 1H), 3.12 (d,
J
= 5.1 Hz, 2H), 1.78
1
3
(d,
J
= 51.3 Hz, 2H), 1.55 – 1.37 (m, 22H). C NMR (101 MHz, CDCl ) 155.90,
δ
3
1
25.01, 80.08, 79.36, 53.32, 40.08, 32.02, 29.56, 28.43, 28.36, 22.42.
1
N-Boc-L-alanine (2e): White solid, yield: 89.4%. H NMR (400 MHz, CDCl )
δ
3
1
3
1
0.29 (s, 1H), 4.26 (d,
MHz, CDCl3) 177.91, 155.46, 80.26, 49.13, 28.31, 18.42.
N-Boc-L-leucine (2f): Colourless viscous oily liquid, yield: 75%. H NMR (400 MHz,
CDCl3) 8.16 (s, 1H), 4.23 (d, = 72.5 Hz, 1H), 1.81 – 1.53 (m, 3H), 1.45 (s, 9H),
J = 72.0 Hz, 1H), 1.44 (d, J = 8.1 Hz, 12H). C NMR (101
δ
1
δ
J

ACCEPTED MANUSCRIPT
1
3
1
2
.00 (t,
J
= 30.9 Hz, 6H). C NMR (101 MHz, CDCl )
δ
155.75, 80.13, 52.08, 41.45,
3
8.29, 24.78, 22.83, 21.76.
1
N-Boc-L-isoleucine (2g): Colourless viscous oily liquid, yield: 78.8%. H NMR (400
MHz, CDCl3)
0H), 1.23 (ddd,
MHz, CDCl3)
δ
8.13 (s, 1H), 4.37 – 3.99 (m, 1H), 1.91 (s, 1H), 1.44 (d,
J
= 8.6 Hz,
13
1
J = 21.1, 13.5, 4.8 Hz, 1H), 1.02 – 0.90 (m, 6H). C NMR (101
δ
177.75, 155.79, 80.06, 57.87, 37.75, 28.30, 24.88, 15.49, 11.60.
1
N-Boc-L-phenylalanine (2h): Colourless viscous oily liquid, yield: 72.2%. H NMR
(400 MHz, CDCl3) δ 7.32 – 7.16 (m, 5H), 6.95 (s, 1H), 4.59 (d, J = 5.6 Hz, 1H), 3.24
1
3
–
1
3.01 (m, 2H), 1.49 – 1.25 (m, 9H). C NMR (101 MHz, CDCl )
δ
175.85, 155.43,
3
36.03, 129.44, 128.56, 127.02, 80.21, 54.37, 37.86 28.30.
1
N-Boc-L-threonine (2i): Colourless viscous oily liquid, yield: 72.2%. H NMR (400
MHz, CDCl3)
Hz, 1H), 1.53 – 1.38 (m, 9H), 1.27 (dd,
CDCl3) 174.31, 156.66, 80.54, 67.82, 58.64, 28.30, 19.29.
δ
5.63 (d,
J
= 8.2 Hz, 1H), 4.70 (s, 3H), 4.40 (s, 1H), 4.27 (d,
J
= 7.5
1
3
J
= 11.5, 5.1 Hz, 3H). C NMR (101 MHz,
δ
General procedure for the synthesis of intermediate molecules (2b):
L-arginine (8.7 g, 50 mmol) was added into a solution of tert-butanol (150 mL) and
water (150mL) in a 500 mL round-bottom flask. The mixture was cooled to 0 °C in an
ice bath and NaOH (7.0 g, 175 mmol) was added. The solution was stirred for 5 min
at 0 °C and to it was added di-tert-butyl dicarbonate (43.7 g, 200 mmol) in
portions.[53] The reaction mixture was stirred for 48 h at room temperature and
monitored by TLC (CH Cl /MeOH/CH COOH, 40:1:0.2). The organic solvent (150
2
2
3
mL) was evaporated under reduced pressure and the residue was extracted with ethyl
ether. The extracted solution was divided into three layers in the separating funnel, the
middle layer was collected and carefully acidified with citric acid to pH 3~4 and was
extracted with ethyl acetate (3 × 60 mL). The extract was dried with anhydrous
sodium sulfate and evaporated in a vacuum. The white solid was obtained after dried
in a vacuum oven.

ACCEPTED MANUSCRIPT
1
Boc-Arg(Boc) -OH (2b): White solid, yield: 61.1%. H NMR (400 MHz, CDCl )
δ
2
3
5
1
1
4
.71 (s, 1H), 4.33 (s, 1H), 3.88 (s, 2H), 3.42 (s, 1H), 1.73 (dd, J = 37.7, 31.2 Hz, 5H),
1
3
.52 (d, J = 5.0 Hz, 9H), 1.49 (s, 9H), 1.44 (s, 9H). C NMR (101 MHz, CDCl ) δ
3
71.29, 156.20, 154.75, 153.17, 84.32, 83.19, 79.71, 79.45, 77.39, 77.07, 76.75, 60.43,
4.35, 40.51, 29.02, 28.36, 28.18, 28.02, 27.96, 27.89, 24.67, 21.02, 14.16.
General procedure for the synthesis of intermediate molecules (2c, 2d): To a
stirred solution of L-serine and D-serine (5.2 g, 50.0 mmol) in 1M NaOH aqueous
solution (50 mL) and 1, 4-dioxane (100 mL), di-tert-butyl dicarbonate (13.1 g, 60.0
mmol) was slowly added at 0 °C. The mixture was warmed to room temperature and
stirred for 24h. After evaporation of 1, 4-dioxane, the aqueous layer was wash with
Et O (50 mL). The aqueous layer was acidified with 1M H SO aqueous solution to
2
2
4
give pH 2-3 and extracted with ethyl acetate (3×50 mL). The combined organic layer
was dried over anhydrous Na SO , filtrated, and concentrated under reduced pressure
2
4
to give the product N-Boc-L-serine.
N-Boc-L-serine (2c): White solid, yield: 53.5%. H NMR (400 MHz, CDCl )
1
δ
6.63
= 77.2
173.93, 156.30, 80.67, 62.99, 55.50, 28.30.
3
(d,
J = 46.7 Hz, 2H), 4.30 (d, J = 60.7 Hz, 1H), 4.16 – 3.80 (m, 2H), 1.36 (d, J
13
Hz, 9H). C NMR (101 MHz, CDCl )
δ
3
1
N-Boc-D-serine (2d): White solid, yield: 53.2%. H NMR (400 MHz, CDCl )
δ
6.04
s, 2H), 4.36 (s, 1H), 4.05 (s, 1H), 3.91 – 3.81 (m, 1H), 1.45 (s, 9H). C NMR (101
MHz, CDCl3) 173.90, 156.28, 80.68, 62.99, 55.47, 28.30.
3
1
3
(
δ
General procedure for the synthesis of amide coupling compounds (3a-3u):
To a stirred solution of N, N-di-n-alkylamines ( 1a-1i) (200mg, 1 equiv) in 5:2
DMF/CHCl (8.4 mL) was added N, N-di-isopropyl-ethylamine (DIPEA, 3 equiv) at
3
0
°C. To this solution was then added HBTU (1.25 equiv). This mixture was stirred
for 5 min at 0 °C, and subsequently the Boc-protected amino acids (2a-2i) (1.25 equiv)
were added to it. The mixture was stirred at 0 °C for 30 min and subsequently at RT
for 24 h typically[39]. At the end, CHCl was evaporated under reduced pressure and
3
the resulting solution was diluted to 2 times its original volume by addition of ethyl

ACCEPTED MANUSCRIPT
acetate. This mixture was subsequently washed with 0.5 M KHSO , H O (three times),
4
2
and brine. The combined organic layer was dried over anhydrous Na SO , filtrated,
2
4
and concentrated under reduced pressure and the residue was purified using column
chromatography (petroleum ether: ethyl acetate= 5:1 v/v and with 5%triethylamine
to obtain the product in 40%-80% yield. The purified compound was subsequently
)
1
13
characterized using H NMR and C NMR.
Di-tert-butyl (6-(diethylamino)-6-oxohexane-1,5-diyl)(S)-dicarbamate (3a):
1
Colourless oily liquid, yield: 51.2%. H NMR (400 MHz, CDCl )
δ 5.40 (d, J = 8.7
3
Hz, 1H), 4.66 (s, 1H), 4.53 (d,
J
= 4.6 Hz, 1H), 3.52 (dd,
= 13.6, 7.0 Hz, 1H), 3.10 (dd,
.48 (m, 4H), 1.43 (s, 18H), 1.41 – 1.35 (m, 2H), 1.23 (t,
J
= 13.6, 7.0 Hz, 1H), 3.43 –
3
1
7
4
.31 (m, 2H), 3.23 (dd,
J
J = 12.8, 6.5 Hz, 2H), 1.70 –
J
= 7.1 Hz, 3H), 1.12 (t,
171.56, 156.05, 155.62, 79.52, 49.85,
1.89, 40.35, 33.68, 28.44, 28.39, 22.52, 14.63, 12.94.
J =
13
.1 Hz, 3H). C NMR (101 MHz, CDCl )
δ
3
Di-tert-butyl (6-(dibutylamino)-6-oxohexane-1,5-diyl)(S)-dicarbamate (3b): Yellow
1
oily liquid, yield: 72.0%. H NMR (400 MHz, CDCl )
δ
5.37 (d,
J
= 8.3 Hz, 1H), 4.67
= 15.1, 8.2 Hz, 2H),
J = 11.5, 7.2 Hz, 2H), 1.59 – 1.47 (m, 6H), 1.43 (s,
3
(s, 1H), 4.53 (d,
J
= 4.3 Hz, 1H), 3.54 – 3.45 (m, 1H), 3.26 (dd,
.17 – 3.03 (m, 3H), 1.63 (dd,
8H), 1.40 – 1.27 (m, 6H), 0.94 (dt,
171.90, 156.03, 155.57, 79.44, 60.39, 49.84, 47.54, 45.76, 33.80, 31.37,
9.72, 28.44, 28.38, 22.56, 20.18, 20.07, 13.85, 13.82.
J
3
1
1
3
J
= 12.6, 7.4 Hz, 6H). C NMR (101 MHz,
CDCl3)
δ
2
Di-tert-butyl (6-(dipentylamino)-6-oxohexane-1,5-diyl)(S)-dicarbamate (3c): Yellow
1
oily liquid, yield: 72.5%. H NMR (400 MHz, CDCl )
δ
5.37 (d,
= 7.7 Hz, 1H), 3.25 (dd,
J = 13.5, 9.2 Hz, 2H), 1.58 – 1.46 (m, 6H), 1.44 (s,
J
= 7.4 Hz, 1H), 4.66
3
(s, 1H), 4.53 (d,
J
= 4.5 Hz, 1H), 3.48 (d,
H), 3.17 – 3.04 (m, 3H), 1.62 (dd,
8H), 1.31 (ddd, = 24.0, 14.3, 7.0 Hz, 10H), 0.90 (dd,
171.90, 156.03, 155.55, 79.44, 49.83, 48.31, 47.75, 45.98,
0.41, 33.82, 29.71, 29.50, 29.12, 28.98, 28.44, 28.38, 22.57, 22.44, 14.03, 13.98.
J
J
= 15.3, 8.2 Hz,
2
1
1
3
J
J
= 14.7, 7.3 Hz, 6H).
C
NMR (101 MHz, CDCl3)
δ
4
Di-tert-butyl (6-(dihexylamino)-6-oxohexane-1,5-diyl)(S)-dicarbamate (3d): Yellow
1
oily liquid, yield: 71.9%. H NMR (400 MHz, CDCl )
δ
5.37 (d,
J
= 8.4 Hz, 1H), 4.65
3
(s, 1H), 4.53 (d,
J
= 4.4 Hz, 1H), 3.53 – 3.44 (m, 1H), 3.25 (dd,
J
= 15.3, 8.6 Hz, 2H),

ACCEPTED MANUSCRIPT
3
1
1
2
.15 – 3.06 (m, 3H), 1.64 (dd,
J = 13.5, 5.1 Hz, 2H), 1.59 – 1.48 (m, 6H), 1.43 (s,
1
3
8H), 1.33 – 1.25 (m, 14H), 0.89 (q,
J
= 6.5 Hz, 6H). C NMR (101 MHz, CDCl ) δ
3
71.88, 156.02, 155.55, 79.44, 49.83, 47.79, 46.01, 40.40, 33.83, 31.55, 31.53, 29.51,
9.27, 28.44, 28.38, 27.54, 26.63, 26.54, 22.59, 14.02, 13.98.
Di-tert-butyl (6-(diheptylamino)-6-oxohexane-1,5-diyl)(S)-dicarbamate (3e): Yellow
1
oily liquid, yield: 62.6%. H NMR (400 MHz, CDCl )
δ
5.37 (d,
J
= 8.3 Hz, 1H), 4.66
3
(
s, 1H), 4.53 (d,
J
= 3.7 Hz, 1H), 3.52 – 3.43 (m, 1H), 3.25 (dd,
.15 – 3.06 (m, 3H), 1.62 (dd,
8H), 1.28 (dd, = 8.4, 6.0 Hz, 18H), 0.87 (dd,
156.01, 155.54, 79.42, 49.82, 47.77, 45.99, 40.38, 33.81, 31.77, 31.73,
9.30, 29.02, 28.42, 28.36, 27.58, 26.92, 26.82, 22.57, 22.54, 14.04.
J = 15.5, 9.0 Hz, 2H),
3
1
J = 13.3, 8.1 Hz, 2H), 1.58 – 1.48 (m, 6H), 1.43 (s,
1
3
J
J = 6.8, 4.1 Hz, 6H). C NMR (101
MHz, CDCl3)
δ
2
Di-tert-butyl (6-(dioctylamino)-6-oxohexane-1,5-diyl)(S)-dicarbamate (3f): Yellow
1
oily liquid, yield: 67.8%. H NMR (400 MHz, CDCl )
δ
5.37 (d,
= 6.6 Hz, 1H), 3.25 (d,
J = 13.1, 8.1 Hz, 2H), 1.58 – 1.47 (m, 6H),
J
= 8.4 Hz, 1H), 4.66
3
(s, 1H), 4.53 (d,
J
= 4.2 Hz, 1H), 3.47 (d,
= 12.0, 6.6 Hz, 3H), 1.62 (dd,
.43 (s, 18H), 1.27 (dd, = 13.5, 6.6 Hz, 22H), 0.91 – 0.85 (m, 6H). C NMR (101
171.87, 156.02, 155.55, 79.43, 49.83, 47.79, 46.01, 40.40, 33.83,
1.80, 31.76, 29.34, 29.25, 29.22, 28.44, 28.38, 27.60, 26.99, 26.89, 22.63, 22.56,
4.09, 14.07.
J
J
= 6.5 Hz, 2H),
3
1
.11 (dd,
J
1
3
J
MHz, CDCl3)
δ
3
1
Di-tert-butyl (6-(dinonylamino)-6-oxohexane-1,5-diyl)(S)-dicarbamate (3g): Yellow
1
oily liquid, yield: 66.8%. H NMR (400 MHz, CDCl )
δ
5.38 (d,
= 15.0, 7.6 Hz, 1H), 3.33 – 3.19 (m,
H), 3.16 – 3.04 (m, 3H), 1.66 – 1.59 (m, 2H), 1.58 – 1.47 (m, 6H), 1.43 (s, 18H),
J = 8.5 Hz, 1H), 4.69
3
(
s, 1H), 4.53 (dd, J = 12.7, 8.3 Hz, 1H), 3.48 (dt, J
2
1
1
2
2
13
.26 (s, 26H), 0.91 – 0.85 (m, 6H). C NMR (101 MHz, CDCl )
δ 171.86, 156.01,
3
55.53, 79.40, 78.95, 49.80, 47.77, 46.00, 40.36, 33.77, 31.82, 31.81, 29.67, 29.51,
9.48, 29.34, 29.28, 29.22, 29.19, 28.41, 28.35, 27.57, 26.95, 26.85, 22.63, 22.61,
2.53, 14.06.
di-tert-butyl (6-(didecylamino)-6-oxohexane-1,5-diyl)(S)-dicarbamate (3h): Yellow
1
oily liquid, yield: 74.0%. H NMR (400 MHz, CDCl )
δ
5.36 (d,
J = 8.6 Hz, 1H), 4.62
3
(d,
J
= 17.5 Hz, 1H), 4.53 (d,
J
= 4.2 Hz, 1H), 3.46 (s, 1H), 3.32 – 3.17 (m, 2H), 3.16