Tyrocidine A Analogues Bearing the Planar d-Phe-2-Abz Turn Motif: How Conformation Impacts Bioactivity

Article abstract of DOI:10.1021/acs.jmedchem.7b00953

The d-Phe-Pro β-turn of the cyclic β-hairpin antimicrobial decapeptide tyrocidine A, (Tyrc A) was substituted with the d-Phe-2-aminobenzoic acid (2-Abz) motif in a synthetic analogue (1). The NMR structure of 1 demonstrated that compound 1 retained the β-hairpin structure of Tyrc A with additional planarity, resulting in approximately 30-fold reduced hemolysis than Tyrc A. Although antibacterial activity was partially compromised, a single Gln to Lys substitution (2) restored activity equivalent to Tyrc A against S. aureus, enhanced activity against two Gram negative strains and maintained the reduced hemeloysis of 1. Analysis by transmission electron microscopy (TEM) suggested a membrane lytic mechanism of action for these peptides. Compound 2 also exhibits nanomolar antifungal activity in synergy with amphotericin B. The d-Phe-2-Abz turn may serve as a tool for the synthesis of structurally predictable β-hairpin libraries. Unlike traditional β-turn motifs such as d-Pro-Gly, both the 2-Abz and d-Phe rings may be further functionalized.

Full text of DOI:10.1021/acs.jmedchem.7b00953

Subscriber access provided by READING UNIV
Article
Tyrocidine A Analogues Bearing the Planar D-Phe-2-
Abz Turn Motif: How Conformation Impacts Bioactivity
Alan J. Cameron, Patrick J.B. Edwards, Elena Harjes, and Vijayalekshmi Sarojini
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.7b00953 • Publication Date (Web): 15 Nov 2017
Downloaded from http://pubs.acs.org on November 16, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 33
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
Tyrocidine A Analogues Bearing the Planar
How Conformation Impacts Bioactivity
D-Phe-2-Abz Turn Motif:
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Alan J. Camerona, Patrick J. B. Edwardsb, Elena Harjesb and Vijayalekshmi Sarojinia,*
aSchool of Chemical Sciences, University of Auckland, Auckland 1142, New Zealand.
bInstitute of Fundamental Sciences, Massey University, Palmerston North 4442, New Zealand.
*Author for correspondence
Vijayalekshmi Sarojini
Phone: + 64 9 9233387
E-mail: v.sarojini@auckland.ac.nz
1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 33
1
2
3
4
Abstract
5
6
7
8
9
The
D
-Phe-Pro β-turn of the cyclic β-hairpin antimicrobial decapeptide Tyrocidine A, (Tyrc A)
-Phe-2-aminobenzoic acid (2-Abz) motif in a synthetic analogue (1).
was substituted with the
D
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
NMR structure of 1 demonstrated that compound 1 retained the β-hairpin structure of Tyrc A
with additional planarity, resulting in approx. 30-fold reduced haemolysis than Tyrc A. Although
antibacterial activity was partially compromised, a single Gln to Lys substitution (2) restored
activity equivalent to Tyrc A against S. aureus, enhanced activity against two Gram negative
strains and maintained the reduced haemloysis of 1. Analysis by transmission electron
microscopy (TEM) suggested a membrane lytic mechanism of action for these peptides.
Compound 2 also exhibits nanomolar antifungal activity in synergy with amphotericin B. The
D-
Phe-2-Abz turn may serve as a tool for the synthesis of structurally predictable β-hairpin
libraries. Unlike traditional β-turn motifs such as
D-Pro-Gly, both the 2-Abz and D-Phe rings
may be further functionalized.
Introduction
Tyrocidine A (Tyrc A, Figure 1) is a membrane lytic cyclic antimicrobial peptide (AMP) with
low micromolar activity against Gram positive pathogens and fungi, but is largely inactive
against Gram negative bacteria.1-3 Like gramicidin S (Figure 1), Tyrc A also forms a cyclic β-
sheet.4 The X-ray crystal structure of Tyrc A reveals the presence of a type II’ β-turn formed by
residues Leu10, D-Phe1, Pro2 and Phe3 with Cα – Cα distance of 5.28 Å.5 At the other end of
i
i+3
the molecule, a distorted type I turn is formed.5 The molecule forms an amphipathic β-sheet with
a curvature (Figure 3A) distinct from the more planar backbone of gramicidin S.5 The
2
ACS Paragon Plus Environment

Page 3 of 33
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
amphipathicity arises from the lower face of the molecule (concave face) containing more
hydrophilic and cationic side chains, while the upper face is predominantly hydrophobic (Figure
3).5 The crystal structure also revealed Tyrc A to occur as an amphipathic intermolecular dimer,
a feature likely important in its antibacterial mode of action.5 Tyrothricin, a mixture of
tyrocidines A-D and gramicidin, was the first commercial antibiotic mixture, used under trade
names Limex and Tyrosur3-4, 6 Although this mixture is noted to have significant antibacterial,
antifungal and antimalarial properties, its clinical use and that of both Tyrc A and gramicidin S is
limited to topical applications due to significant haemloysis and toxicity to both liver and
kidneys.2-4, 7 Additionally, this family of peptides possess poor activity against Gram-negative
bacteria.8-9. Overcoming these major drawbacks may pave the way towards their development as
broad spectrum antimicrobials.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Unique turn conformations of 2-amino benzoic acid (2-Abz) containing peptides have previously
been reported in model peptides devoid of known biological activities.10-13 Work from our
laboratory demonstrated that
D
-Phe-2-Abz motif adopts a unique turn conformation.14 We
-Phe-2-Abz as a turn motif in the synthesis of
wanted to explore the applications of
D
antimicrobial peptides, the development of which is a major focus of our research.15-16 Tyrc A,
with its well characterised β-hairpin structure and antimicrobial activity, but inherent drawbacks,
appeared a good candidate for these studies. The relationship between the molecular
conformation and bioactivity of two novel analogues (Figure 1), employing the
turn motif, are presented, in comparison to Tyrc A.
D-Phe-2-Abz
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 33
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1: (A) Amino acid sequences of Tyrc A, gramicidin S and, compounds 1and 2. D-Amino
acids are represented by lower case letters. 2-Abz represents 2-aminobenzoic acid and O
represents ornithine. Bold letters are used to denote amino acids altered in the analogues. (B)
Crystal structure of Tyrc A, showing hydrogen bonds; taken from PDB ID: 4M6E.5
Results and Discussion
The turn nucleated by D-Phe1-Pro2 in Tyrc A was replaced with the
D-Phe-2-Abz turn motif. The
additional backbone atom in 2-Abz and rigidity imparted by this residue was expected to result
in an altered molecular architecture, allowing us to gain new insights as to how the conformation
influences the biological properties of the molecule. Antibacterial activity was investigated
against E. coli, P. aeruginosa and S. aureus by minimal inhibitory concentration (MIC) analyses
and toxicity measured by haemolysis of mouse red blood cells (RBCs). The novel Tyrc A
4
ACS Paragon Plus Environment

Page 5 of 33
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
analogue, compound 1 (Figure 1), was approximately 30-fold less haemolytic than Tyrc A
(Figure 2). However, its antibacterial activity against S. aureus was moderately reduced (four-
fold, Table 1).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1: Antibacterial activity of Tyrc A and analogues 1-2
MIC (µM)a
Compound
E.coli
P. aeruginosa
S. aureus
1.56
Tyrc A
25
100
100
25
50
6.25
1
2
12.5
1.56
a MIC determined from three independent experiments each with three internal replicates.
140
120
100
80
60
40
20
0
0
10
20
30
40
50
60
70
80
90
100
-20
Peptide concentration (µM)
5
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 33
1
2
3
4
5
6
7
8
9
Figure 2: Percentage of haemolysis of mouse blood cells at various concentrations of Tyrc A
and compounds 1 and 2. 100% haemolysis referenced as treatment with 0.5 % triton X-100.
Inset: percentage haemolysis plotted against log peptide concentration (µM).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Compound 1 was poorly soluble in water and therefore a 1:1 mixture of CD3CN and H2O was
chosen for nuclear magnetic resonance (NMR) studies. Well dispersed resonances were observed
in this solvent mixture (Figure S3; Table S2). Additionally, this solvent system may better mimic
the interaction of the peptide with biological membranes than aqueous conditions. Strong Cα H –
i
3
Ni+1H rotating frame Overhauser effects (ROEs) and large JNH-C H
α
coupling constants, ranging
from 7.3 - 8.8 Hz were observed for the β-strand residues (Phe3-Asn5 and Val8-Leu10),
supporting an extended backbone conformation typical of β-sheets. A diagram highlighting key
inter-strand ROEs observed for the cyclic β-hairpin 1 is shown in Figure S4. Structures were
calculated with experimentally derived ROE restrains using the standard NMR structure
calculation protocol of the YASARA molecular modelling package. The ensemble of calculated
structures showed a conformational shift, for compound 1, to a more planar architecture than
Tyrc A (Figure 3). The increased planarity sees the hydrophilic Tyr7 residue drifting towards the
hydrophobic upper face, while the hydrophilic Asn5 residue also shifts and now resides directly
on the upper face (Figure 3). Furthermore, the Phe3 side chain shifts away from the hydrophobic
face towards the lower face which is predominantly hydrophilic for Tyrc A. All nine ensemble
members demonstrated a similar geometry with a root mean square deviation (RMSD) average
of only 0.98 Å for the total structure (Figure 4). Both the increased planarity of the backbone and
side chain shifts, especially that of Phe3, result in a reduction in the amphipathic organisation of
the molecule and in-turn, appear responsible for the reduced haemolysis of compound 1. In
6
ACS Paragon Plus Environment

Page 7 of 33
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
support of the reduced hydrophobic upper face presented by compound 1, the reversed-phase
high performance liquid chromatography (RP-HPLC) retention time of this compound was
reduced approx. 5 minutes compared to Tyrc A even on a gradient with reduced solvent B (Table
S1).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Given the Gln6 side chain is oriented towards the cationic face in both 1 and Tyrc A (Figures 3
and 4) and is not involved in stabilising the intramolecular structure or dimerization of Tyrc A5,
it appeared a good candidate for substitution in order to improve the antibacterial activity of 1.
In-fact, a simple point substitution of Gln6 for lysine in Tyrc A was reported by Marquez et al8.
While they found this substitution to enhance the already potent activity against Gram positive
bacteria, little or no advantage was achieved for the Gram negative E.coli with MICs >100 µM.
Effect of this substitution on haemolysis was not reported.
In an attempt to improve the antibacterial activity of 1, we synthesised compound 2 where Gln6
in 1 was substituted with lysine (Figure 1). To our delight we found that the reduced haemolysis
of 1 (approx. 30 fold) was maintained by compound 2 (Figure 2) while the spectrum of activity
was broadened and potency enhanced (Table 1). Activity of 2 against S. aureus was equivalent to
that of Tyrc A (MIC of 1.56 µM) and also increased by two and four fold against E. coli and P.
aeruginosa and respectively (Table 1).
7
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 33
1
2
3
4
5
6
7
8
9
The NMR structure of compound 2 (NMR parameters given in Table S3) demonstrated a
structural shift from Tyrc A to a more planar architecture nearly identical to that of compound 1
(Figure 3). The structural similarity is supported by the observation of many key intra- and inter-
stand ROEs common to both compounds 1 and 2 (Figures S4-5). The ensemble of eight
structures for compound 2 was also well defined (Figure 4), demonstrating an RMSD average of
only 1.12 Å for the total structure. The conserved structure of compound 2, relative to its
predecessor, compound 1, supports our observation that the D-Phe-2-Abz turn is responsible for
the reduction in haemolysis of compounds 1 and 2, by means of increased ring planarity and side
chain shifts that ultimately disrupts amphipathicity and hydrophobicity of the upper face (Figures
3 and 4). In further support of this argument, Gln6 substitution for ornithine (Orn) or Lys in Tyrc
A, documented by Qin et al17, provided much lesser reductions in haemolysis (3 or 12-fold
respectively) than observed for compounds 1 and 2 (approx. 30-fold). Similar observations on
amphipathic disruption have been previously reported for designed AMPs whereby, substitutions
leading to a less well-defined amphipathic character resulted in reduced haemolysis.18 The
enhanced activity of 2 (compared to 1) is then attributable to the increased cationic charge of the
lower face where both Lys6 and Orn9 now reside (Figures 3, and 4). This is in accordance with
previous Tyrc A analogues, in which point substitution of cationic residues on the hydrophobic
upper face saw reduced haemolysis, while their introduction to the lower face predominantly
increased antibacterial activity.8, 19-20 The crystal structure of Tyrc A appears to accurately
represent the bioactive conformation in solution, as it supports the results of previous structure-
activity-relationship studies based on point substitutions.5, 8 While the changes in amphipathicity
for the calculated structures of compounds 1 and 2 account for both the observed changes in
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
8
ACS Paragon Plus Environment

Page 9 of 33
Journal of Medicinal Chemistry
1
2
3
4
bioactivity and reduced HPLC retention time of 1, it should be noted that there is of course
5
6
potential for discrepancy when comparing NMR and crystal structures.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3: Left; Perspective views showing hydrogen bonds, for the crystal structure of Tyrc A
(A) and single representative NMR structure calculated for 1 (B) and 2 (C). Right; the backbone
of each molecule is shown with side chains removed to highlight the curvature of Tyrc A (D) and
the greater planarity of both 1 (E) and 2 (F). The crystal structure of Tyrc A was taken from PDB
ID: 4M6E.5
9
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 33
1
2
3
4
5
6
7
A
B
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4: Ensemble of lowest energy NMR structures calculated for compound 1 (A) and
compound 2 (B) showing nine and eight structures respectively. The predicted 3JNH-C H
α
values of
the calculated structures were generally in good agreement with the experimentally determined
values (Tables S2 and S3).
While previous efforts by Kohli et al19 to enhance the therapeutic index of Tyrc A by making
point substitutions provided two compounds with significantly reduced haemolysis, the activity
of both was compromised against Gram positive bacterial strains (approx. four and nine-fold
against MRSA) and remained poor or moderate respectively against two Gram negative
pathogens, E. coli (120 µM and 290 µM ) and P. aeruginosa (30 µM and 37 µM). On the other
hand, compound 2 reported here, possessing an altered turn framework and molecular
architecture, reduces haemolysis approx. 30-fold while also enhancing activity against Gram
negative species and retaining potency against Gram positive S. aureus. Thus, the improved
bioactivity properties of compound 2 highlights the potential therapeutic applications of peptides
employing the D-Phe-2-Abz turn motif.
10
ACS Paragon Plus Environment

Page 11 of 33
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
As stated above, the β-turn nucleated by the
D
-Phe-2-Abz motif is almost identical for
8
9
compounds 1 and 2 (Figure 3). This β-turn structure demonstrated similar Cα – Cαi+3 distances to
i
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Tyrc A and is shown for compound 2 alongside the turn structure of Tyrc A (Figure 5).
Crucially, the full ensemble of compound 2 (Figure 4) demonstrated appropriate Cα – Cα
i
i+3
distances of 5.39 – 5.78 Å satisfying the 7 Å limit used for defining β-turns.21-23 The
D
-Phe-2-
Abz motif demonstrates an intra-residue hydrogen bond (H-bond) of 2-Abz which stabilises the
planar conformation and appears to constitute a minimal β-turn motif (Figures 3 and 5).
Stabilisation of the 2-Abz geometry by this intra-residue H-bond has previously been observed in
both the crystal and solution states by the Sanjayan group.10-11 Despite both compounds 1 and 2
maintaining a β-turn with a similar Cα – Cα distance to native Tyrc A (Figure 5), a shift in
i
i+3
geometry about Leu10, Phe3 and the N-terminus of -Phe4 sees only one of the three inter-strand
D
backbone H-bonds of Tyrc A maintained, between Asn5 NH and the carbonyl of Val8 (Figure 3).
However, despite the reduction in inter-strand H-bonds, the well-defined (low RMSD)
ensembles of both analogues suggest a robust and rigid framework is maintained in the presence
of the D-Phe-2-Abz turn motif. At the other end of the molecule, the bifurcated H-bond of the
Asn5 side chain carbonyl to the NH of both Tyr7 and Val8 is maintained similarly to Tyrc A,
despite the increased backbone planarity of these analogues (Figure 3). This side chain-backbone
interaction appears to stabilise both the β-turn and side chain conformation at this end of the
molecule (Figure 4). Additionally, like Tyrc A, ensemble members of both analogues
demonstrated possible backbone H-bond contacts for the Orn9 side chain on the lower face of the
molecules (not shown). The H-bonds observed in the modelled structures of compounds 1 and 2
(Figure 3) are further supported by low amide temperature coefficients (dδ/dT) of 2-Abz2, Asn5,
11
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 33
1
2
Tyr7 and Val8, ranging -2.4 to -4.2 ppb/K, compared to that of
D
-Phe1 (-7.6 and -7.3 ppb/K)
3
4
5
6
7
8
which was not predicted to form any H-bond contacts and appears solvent exposed.24-25 The
temperature coefficient of Leu10 was also surprisingly low (approx. -1 ppb/K). This could be
explained by a weaker H-bond occurring between the NH of Leu10 and the carbonyl of 2-Abz-2
(Figure 5). For compound 2 donor – acceptor distances ranging 3.4 – 3.8 Å were observed for
this interaction in all eight ensemble members, while for compound 1, this interaction appears
less likely with donor – acceptor distances slightly > 4 Å. The occurrence of this i – i+2 H-bond
is in accordance with our previous observations from the crystal structure of the model β-turn
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
tetrapeptide D-leu-D-Phe-2-Abz-D
-Ala.14 The temperature coefficients of the Gln6 / Lys6 amides
were also relatively low, - 3.9 and -4.3 ppb/K for compounds 1 and 2 respectively. Based on the
modelled structures, it appears unlikely this is due to any H-bonding interaction and instead may
be due to solvent shielding by the Asn5 side chain. Similar effects of side chains have been
previously observed for cyclic peptides and may also contribute to the low temperature
dependence of Leu10.24, 26
Figure 5: (A) Representative molecule from the ensemble of 2 showing β-turn structure induced
by the
D
-Phe-2-Abz motif. The intra-residue H-bond of 2-Abz2 and a likely i – i+2 H-bond
12
ACS Paragon Plus Environment

Page 13 of 33
Journal of Medicinal Chemistry
1
2
-Leu10 amide and carbonyl of 2-Abz2 are also shown. (B) Crystal structure of Tyrc
3
4
between the
D
5
6
A showing the β-turn induced by the
D-Phe-Pro motif, with the alternate Leu conformation
7
8
9
removed. The Cα – Cαi+3 distance of each (A and B) is highlighted. The crystal structure of Tyrc
i
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
A was taken from PDB ID: 4M6E.5
The antibacterial mode of action of compound 2 and synthetic Tyrc A was investigated by TEM
against both S. aureus and P. aeruginosa (Figures 6 and 7). S. aureus control cells showed
relatively uniform and intact cell membranes and cell walls. The cell wall was noted to be
slightly diffuse in areas, particularly for dividing cells (Figure 6, right).27 The intracellular area
consistently demonstrated high density and in some sections, particularly dividing cells, genetic
material (nucleoid) was clearly seen.27-29 However, S. aureus cells with ruptured cell membranes
and cell walls were consistently observed after one hour treatment with both Tyrc A and
compound 2.5 Peptide treated cells consistently showed reduced intracellular density, suggesting
cellular leakage, even when the section did not reveal obvious membrane damage (Figure 6,
right). The observations from the TEM images are consistent with those of other studies
employing either transmission or scanning electron microscopy (SEM).15-16, 30-32
13
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 33
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6: TEM micrographs of S. aureus untreated (control) or treated with Tyrc A and
compound 2.
14
ACS Paragon Plus Environment

Page 15 of 33
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
P. aeruginosa control cells consistently demonstrated smooth in-tact cell membranes. Staining of
the intracellular space was less consistent than for S. aureus, particularly towards the centre of
the cells where the stains may not have penetrated (Figure 7). Despite treatment with Tyrc A at 4
X MIC, P. aeruginosa cells failed to demonstrate any obvious membrane disruption. The cells
did however demonstrate enhanced uptake of the stain, suggesting increased membrane
permeability. This appears consistent with the poor potency (high MIC) of Tyrc A against this
pathogen. On the other hand, cells treated with compound 2 (4 X MIC) consistently
demonstrated blebbing and total membrane lysis. Again, this is consistent with both SEM and
TEM studies of AMPs previously reported.15-16, 30-32
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
15
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 33
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 7: TEM micrographs of P. aeruginosa untreated (control) or treated with Tyrc A and
compound 2.
16
ACS Paragon Plus Environment

Page 17 of 33
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Tyrc A is strongly suggested to interact with bacterial membranes as a dimer which accentuates
the amphipathic structure.5, 33 As reported previously by Munyuki et al33, intermolecular dimers
were observed for TyrcA by electrospray ionization mass spectrometry (ESI-MS), with similar
observations occurring for both compounds 1 and 2 (Figure S2). This data suggested compounds
1 and 2 may interact with bacterial membranes by a mechanism similar to Tyrc A. The presence
of a doubly charged dimeric species in the ESI-MS spectra raised the question as to whether
cyclodimerisation had occurred during synthesis. However, mass spectrometry analysis by
matrix-assisted laser desorption/ionization time of flight (MALDI-TOF) did not detect the
dimeric species (data not shown) and presence of the cyclic monomer was further confirmed by
NMR spectroscopy for 1 and 2.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Antifungal susceptibility testing of compound 2 and Tyrc A was performed against C. albicans
(Table 2). Tyrc A is known to possess membrane lytic activity against fungi, however evidence
has strongly suggested additional antifungal targets.2-3 Tyrc A is also known to act
synergistically with amphotericin B.3 Tyrc A demonstrated an MIC of 3.125 µM, while
compound 2 showed an MIC of 12.5 µM. However, in the presence of sub-MIC concentrations
(0.39 µM and 0.78 µM) of amphotericin B, known to induce pores in fungal membranes,
compound 2 displayed a synergistic effect with potency equivalent to that of Tyrc A, with both
showing MICs in the nanomolar range against C. albicans (Table 2). Presumably this occurs by
either a cooperative membrane lytic mechanism or by enhanced membrane permeability, induced
by amphotericin B, promoting access to additional intracellular targets. While highly effective,
17
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 33
1
2
3
4
5
6
7
8
9
clinical use of amphotericin B is limited by nephrotoxicity, primarily due to a lack of selectivity
between fungal and mammalian membranes.34-35 Therefore the development of synergistic
antifungal strategies allowing lower and less toxic doses of amphotericin B is of great value.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 2: Antifungal activity of Tyrc A and compound 2, showing synergism with Amphotericin
B.
MIC (µM)a
Compound
C. albicans
3.125
C. albicans + 0.39 µM AMBb
C. albicans + 0.78 µM AMBb
Tyrc A
2
0.78
0.78
N/A
0.39
0.39
N/A
12.5
Amphotericin B
1.56
a MIC determined from three independent experiments each with three internal replicates.
b Media supplemented with sub-MIC concentration of amphotericin B.
Conclusions
NMR structures of compounds 1 and 2 demonstrated that the
D-Phe-2-Abz motif can be used as
a minimal β-turn motif replacing the
D
-Phe-Pro β-turn of Tyrc A, maintaining similar Cα – Cα
i
i+3
distances. Compound 2 is a novel Tyrc A analogue that displays a broader spectrum of
antibacterial activity with increased potency against Gram negative pathogens, approximately
30-fold reduced haemolysis and antifungal activity in the nanomolar range when used in synergy
with amphotericin B. Although the biological properties of compound 2 are significantly more
18
ACS Paragon Plus Environment

Page 19 of 33
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
favourable than those of the parent Tyrc A, further fine tuning of the structure would be required
before real therapeutic value can be realised. Nevertheless, both compounds 1 and 2 provide
novel insights into the relationship between the molecular conformation and bioactivity of Tyrc
A, whereby substitution of proline for 2-Abz lead to significantly altered physical and biological
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
properties due to a shift in the molecular architecture. The D-Phe-2-Abz turn motif presents a
framework that is likely to accommodate different amino acids with varied side chain
characteristics and stereo-configurations in the i and i+3 positions for the generation of β-hairpin
libraries, much like the common
motifs such as -Pro-Gly, potential for ring substitutions on 2-Abz greatly expands the scope to
display further functionalities or create conjugates without side chain flexibility. The -Phe ring
-Tyr or more complex
D-Pro-Gly turn. Furthermore, compared with existing turn
D
D
may also anchor additional functional groups via substitution with
structures.
D
Experimental
Chemicals and Reagents
All Fmoc-amino acids, 1-(bis-(dimethylamino)methylene)-1H-1,2,3-triazolo[4,5-b]-pyridinium
3-oxide hexafluorophosphate (HATU) and (3-Hydroxy-3H-1,2,3-triazolo[4,5-b]pyridinato-O)tri-
1-pyrrolidinyl-phosphorushexafluorophosphate PyAOP were purchased from AK scientific,
USA. 2-chlorotritylchloride resin was purchased from Peptides International, USA. N,N-
diisopropylethylamine (DIPEA), piperidine, and other common solvents were purchased from
Sigma Aldrich, NZ. Mueller Hinton broth (MHB, non-cation adjusted) of BD-Difico brand was
purchased from Fort Richard Laboratories, New Zealand.
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 33
1
2
3
4
Peptide Synthesis and Purification
5
6
7
8
9
Linear peptides were synthesised on the 2-chlorotrityl chloride resin following standard Fmoc
solid phase peptide synthesis protocols at 0.1 mmol scale using N,N-dimethylformamide (DMF)
as the solvent. Amino acid side chains were protected as follows, asparagine: trityl (Trt),
glutamine: trityl (Trt), lysine: tert-butyloxycarobonyl (Boc), ornithine: tert-butyloxycarobonyl
(Boc), tyrosine: tert-butyl (tBu). Fmoc deprotections were performed using 20 % piperidine in
DMF twice for 10 minutes each using a large excess. Each coupling reaction was performed
using HATU (144.4 mg, 0.38 mmol, 3.8 eq.) and the respective amino acid in 4-fold excess with
DIPEA (175 µl, 1 mmol, 10 eq.) as the base. Couplings to 2-Abz were performed for 3 h.
Protected linear peptides were cleaved from the resin using 1 % trifluoroacetic acid (TFA) in
dichloromethane (DCM, 10 mL). TFA cleavage was performed repeatedly for five times of 1
min duration each with washings collected into a flask already containing triethylamine (1.5
molar excess to the total TFA). Crude linear peptides were purified to > 95 % purity by RP-
HPLC. Percentage yields of the linear peptides after purification were 11.5 (1), 15.2 (2) and 45
(TyrcA) respectively. RP-HPLC was undertaken using a GE Pharmacia ÄKTA purifier 10
system with a Phenomenex Luna 5 ꢀm C18 100 Å (250 mm × 21.2 mm) column, using H2O with
0.1% TFA as solvent A and 99 % MeCN 1% H2O with 0.1 % TFA at a flow rate of 10 mL/min
with UV detection at 214, 254 and 280 nm (Table S1). Purified peptides were obtained as white
fluffy solids after lyophilisation. For cyclisation, the protected linear peptides were dissolved in
DMF at 1.25 mM with 1 % v/v of DIPEA. After stirring for 5 min, PyAOP (3 eq.) was added and
the reaction allowed to proceed for 4 h before removal of the solvent under vacuum. The crude
cyclic peptides were deprotected using TFA:H2O:TIS cocktail (95:2.5:2.5) for 3 h. The crude
cyclic product was purified to > 95 % purity by RP-HPLC using a linear gradient of H2O and
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
20
ACS Paragon Plus Environment

Page 21 of 33
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
MeCN. Purified cyclic peptides were obtained as white solids after lyophilisation. Purity of
peptides was confirmed by analytical RP-HPLC (Figure S1) using a Phenomenex Luna 5 ꢀm
C18 100 Å (250 mm × 4.6 mm) column and 0.5 mL/min flow rate. The identity of the purified
peptides was confirmed using electrospray ionization mass spectrometry (ESI-MS) recorded on a
Bruker micrOTOFQ mass spectrometer (Figure S2). ESI-MS demonstrated a small peak
corresponding to the intermolecular dimer for compounds 1, 2 and Tyrc A, in accordance with
literature.33 However cyclodimers were not detected after dilution and were absent in MALDI-
MS.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
NMR Spectroscopy
Compounds 1 and 2 were dissolved in a 1:1 mixture of CD3CN (Cambridge Isotopes, 99.8%)
and MQ-H2O at 0.7 mM and 0.8 mM respectively. NMR spectra were recorded on a 5 mm, TCI
cryoprobe-equipped with Bruker Avance 700 spectrometer operating at 700.13 MHz. Peak
assignments were made using 2D double-quantum filtered COSY and TOCSY spectra (mixing
time 80 ms). Distance restraints were obtained from ROESY spectra (mixing time150 ms and
relaxation delay of 3 s). All 2D NMR spectra were recorded at 298 K. 1D 1H NMR spectra were
obtained at 5 K intervals between 288 K and 318 K in order to establish the temperature
dependence of the NH chemical shifts. Spectra were referenced relative to an external standard
of the solvent spiked with TMS at 0 ppm and measured at each temperature.
Spectra were processed with Bruker Topspin 2.1.8 using standard parameters. Peak assignments
and generation of ROE distance restraints was performed using CCPN Analysis v.2.4.2.36 Peak
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 33
1
2
3
4
volumes were calculated using standard settings and ROE distances were calibrated by setting
5
6
7
the average volume of the D-Phe Hβ protons to 1.8 Å.
8
9
Solution Structure Calculation
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Structures of compounds 1 and 2 were calculated using the standard NMR structure calculation
protocol of the program YASARA37 as described.38 In short, the ROE restraints were exported
from CCPN Analysis in XPLOR format and included in structure calculations in all steps.
Calculations additionally included dihedral angle restraints deduced from coupling constants
(3JNH-C H
α
). The dihedral angles for phi were calculated using the Karplus equation.39 As can be
3
seen in the angular dependence of JNH-CαH values published by Pardi et al40, there is a clear
maximum in the distribution at phi = -120° corresponding to J = 9 Hz. This means for J ≥ 9 Hz
3
phi is well defined. However, for the smaller JNH-C H
α
values, phi can be a within a range, such
as for example -120 40° for J = 7 Hz. Therefore, we defined our phi angles loosely as phi = -120
(25-35)° depending on the observed coupling constant and as phi = -60 20° for residue 6
3
(observed JNH-C H
α
= 4.51 or 4.29) in both compounds. As an extra control, we first calculated
structures based only on distance restraints. The introduction of dihedral angles restraints did not
result in ROE violations nor did it change the secondary structure significantly. A hydrogen bond
between two strands was set according to the ROE pattern for both compounds 1 and 2
(Supplementary Figures S4 and S5). First, ten roughly folded structures were calculated by
randomly changing torsion angles starting from the initial structure built with D-amino acids and
non-standard amino acids. The structures were refined in vacuum by running restrained
molecular dynamic simulations in the NOVA force field.37 The protocol involved 40 cycles of
heating and cooling (simulated annealing). Afterwards, to improve the quality of the structures,
22
ACS Paragon Plus Environment

Page 23 of 33
Journal of Medicinal Chemistry
1
2
3
4
they went through water refinement using the same restraints. The structures with the lowest
5
6
energy and retaining
D-configuration (where appropriate) were chosen from 10 structures to
7
8
9
represent the ensemble for each compound. We observed only one ROE from the possible
dimerization of compound 1. This is the ROE between Tyr 7 and Phe 1 and corresponds to the
dimeric interaction observed in Tyrc A crystal structure.5 This ROE was excluded from
calculation, as it was conflicting with other ROEs in the derived monomer structures. We did not
observe other dimeric ROEs, consistent with the expectation of the dimer being less stable in
solution than in the crystal structure. Due to the lack of dimeric ROEs, only monomeric
structures could be calculated. Additionally, the expected dimeric ROEs from Tyrc A
dimerization, are incompatible with a monomeric structure. Exclusion of the single observed
dimeric ROE from the calculation resulted in structures consistent with the majority of ROEs.
Additional data pertaining to structure calculation can be found in Table S4.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Bacterial Minimal Inhibitory Concentration Assay (MIC)
o
S. aureus, E. coli and P. aeruginosa were incubated at 37 C overnight in MHB (non-cation
adjusted). The overnight bacterial cultures were adjusted to an OD of 0.1 and further diluted
1:100 in accordance with a documented literature protocol.41 Assays were performed on 96-well
microtitre plates preparing dilutions in the plates prior to bacterial inoculation. Plates containing
50 ꢀl of the compounds at twice the final test concentrations (or media alone for the non-treated
growth control) were inoculated with 50 ꢀl of the diluted bacterial suspension. The sterility
o
control contained just 100 ꢀl of the media. After incubation at 37 C for 18 h, optical density at
600 nm was measured using a plate reader. MIC was defined as the lowest concentration in
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 33
1
2
3
4
5
6
which no bacterial growth was observed as determined from three concordant experiments, each
with three internal replicates.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fungal Minimal Inhibitory Concentration Assay (MIC)
Fungal susceptibility testing was performed following a literature procedure with minor
changes.42 A single colony of C. albicans was transferred to a solution of Roswell Park
Memorial Institute (RPMI) media (0.5 ml) and adjusted to an OD of 0.1 and then diluted 1:100.
The assay was conducted as per bacterial MIC assays with the incubation period increased to 48
hours. The MIC determination was performed by visual inspection as the fungus does not grow
as a homogeneous solution for plate reading; instead colonies on the bottom of the wells are
observed. MIC was defined as the lowest concentration in which no fungal growth was observed
as determined from three concordant experiments, each with three internal replicates. Synergistic
assays were performed, using media supplemented with amphotericin B (0.78 or 1.56 µM) with
the various compound dilutions in the plate resulting in a final amphotericin concentration of
0.39 and 0.78 µM in each well of the plates.
Haemolysis Assay
Haemolysis of the peptides was tested by determining the haemoglobin release from erythrocyte
suspensions of fresh mouse blood (2 %, v/v). Blood was collected under the University of
Auckland’s Animal Ethics Committee’s approval titled “Animal tissue and blood harvesting,
Production of antisera and/or antibodies” for scientific investigations. The strain of mouse used
24
ACS Paragon Plus Environment

Page 25 of 33
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
was ICR - CD-1. The blood cells were centrifuged for 5 min and the plasma was removed. The
blood cells were washed in Tris buffer (10 mM Tris, 150 mM NaCl, pH 7.2) and the pellet re-
suspended in the Tris buffer (2 % v/v). A two-fold dilution series of the peptides in Tris buffer
was made in the 96-well microtitre plate. To these peptide solutions (50 µL) were added re-
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
o
suspended blood cells (50 µL) and the plates incubated without shaking at 37 C for 1 h. Neat
buffer and 0.5 % Triton X-100 (50 µL) served as the controls. The experiment was performed in
triplicates. After 1 hour, the plates were centrifuged at 3500 x g for 10 min. The supernatant (80
µL) from each well was transferred into fresh 96 well plates and absorbance at 540 nm
measured. Percentage haemolysis at a given concentration of the peptide was calculated from the
following equation, where Aexp is the experimental A540 measurement, ATris is the negative
control where only Tris buffer was added to RBC, and A100%, is the positive control where 0.5 %
Triton X-100 (50 µL) was used to cause lysis of 100 % RBC present.
% haemolysis = (Aexp − ATris)/(A100% − ATris)× 100
Transmission Electron Microscopy (TEM)
S. aureus and P. aeruginosa were grown as overnight cultures in MHB (non-cation adjusted).
Bacterial suspensions were diluted to an OD of 0.4 in MHB. A 1 mL aliquot of each bacterial
suspension was used for each sample preparation. Control samples were incubated for one hour
at 37 oC, while the peptide treated samples were incubated for one hour with each compound at 4
x MIC. After incubation the cells were centrifuged and washed thrice at 5000xg for 3 min with
Tris buffer (10 mM Tris, 150 mM NaCl, pH 7.4). The pellets were then re-suspended in a 4%
gluteraldehyde solution in the buffer and incubated at room temperature for 2 hours. The fixed
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 33
1
2
3
4
5
6
7
8
9
cells were then centrifuged and washed thrice at 5000xg for 3min with Tris buffer. The washed
cells were post-fixed with 1% osmium tetroxide in PBS for 1 hour. The post-fixed cells were
then dehydrated through a graded ethanol series (50, 90 and 100%) before placing in absolute
acetone twice for 10 min each. The supernatant was removed and replaced with a 1:1 mixture of
absolute acetone and epoxy resin for 1 h before transfer to pure epoxy resin for overnight. The
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
o
pellets were then transferred into appropriate tubes and incubated at 60 C for 48 h before
ultrathin sections were obtained using an ultramicrotome. The sections were post-stained with
uranyl acetate and lead citrate. Samples were viewed by transmission electron microscopy (FEI
Technai 12).
Acknowledgments
We thank Stefan Harjes for help with YASARA. AC thanks the University of Auckland and
Freemasons New Zealand for Doctoral Scholarships.
Abbreviations
2-Abz, 2-aminobenzoic acid (anthranilic acid); AMP(s), antimicrobial peptide(s); DBU, 1,8-
Diazabicyclo[5.4.0]undec-7-ene; DCM, dichloromethane; DMF, N,N-dimethylformamide;
DIPEA, N,N-diisopropylethylamine; ESI-MS, electrospray ionization mass spectrometry; Fmoc,
9-fluorenylmethyloxycarbonyl;
H-bond,
hydrogen
bond;
HATU,
1-(bis-
(dimethylamino)methylene)-1H-1,2,3-triazolo[4,5-b]-pyridinium 3-oxide hexafluorophosphate;
MALDI-TOF, matrix-assisted laser desorption/ionization time of flight; MeCN, acetonitrile;
26
ACS Paragon Plus Environment

Page 27 of 33
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
MHB, Mueller-hinton broth; MIC, minimal inhibitory concentration; MRSA, methicillin
resistant Staphylococcus aureus; 2-NBS-Cl, 2-Nitrobenzenesulfonyl chloride; NMP, N-
methylpyrrolidinone; Orn, ornithine; RBC, red blood cells; ROE, rotating frame Overhauser
effect; RMSD, root mean square deviation; RP-HPLC, reversed phase high pressure liquid
chromatography; RPMI, Roswell Park Memorial Institute SEM, scanning electron microscopy;
TEM, transmission electron microscopy; TFA, trifluoroacetic acid, TIS, triisopropyl silane; Tyrc
A, tyrocidine A.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the ACS Publications website.
Physical characteristics of peptides, analytical HPLC traces, ESI-MS spectra, NMR parameters
and spectra, PDB files for compounds 1 (PDBID: 6B34) and 2 (6B35)
SMILES Molecular formula strings (csv file)
Coordinates and experimental data for the compounds 1 and 2 have been deposited with the
Protein Data Bank (www.rcsb.org/pdb) under accession codes 6B34 and 6B35 respectively.
Authors will release the atomic coordinates and experimental data upon article publication.
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 33
1
2
3
4
5
6
7
8
9
References
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1.
2.
3.
Bu, X.; Wu, X.; Xie, G.; Guo, Z. Synthesis of Tyrocidine A and Its Analogues by
Spontaneous Cyclization in Aqueous Solution. Org. Lett. 2002, 4 (17), 2893-2895.
Rautenbach, M.; Troskie, A. M.; Vosloo, J. A.; Dathe, M. E. Antifungal membranolytic
activity of the tyrocidines against filamentous plant fungi. Biochimie 2016, 130, 122-131.
Troskie, A. M.; Rautenbach, M.; Delattin, N.; Vosloo, J. A.; Dathe, M.; Cammue, B. P.
A.; Thevissen, K. Synergistic Activity of the Tyrocidines, Antimicrobial
Cyclodecapeptides from Bacillus aneurinolyticus, with Amphotericin B and Caspofungin
against Candida albicans Biofilms. Antimicrob. Agents Chemother. 2014, 58 (7), 3697-
3707.
4.
5.
Robinson, J. A. β-Hairpin Peptidomimetics: Design, Structures and Biological Activities.
Acc. Chem. Res. 2008, 41 (10), 1278-1288.
Loll, P. J.; Upton, E. C.; Nahoum, V.; Economou, N. J.; Cocklin, S. The high resolution
structure of tyrocidine A reveals an amphipathic dimer. Biochim. Biophys. Acta 2014,
1838 (5), 1199-1207.
6.
7.
8.
9.
Joo, S. H. Cyclic Peptides as Therapeutic Agents and Biochemical Tools. Biomol. Ther.
2012, 20 (1), 19-26.
Rammelkamp, C. H.; Weinstein, L. Toxic effects of tyrothricin, gramicidin and
tyrocidine. J. Infect. Dis. 1942, 71 (2), 166-173.
Marques, M. A.; Citron, D. M.; Wang, C. C. Development of Tyrocidine A Analogues
with Improved Antibacterial Activity. Bioorg. Med. Chem. 2007, 15 (21), 6667-6677.
Gilliver, K. The Inhibitory Action of Antibiotics on Plant Pathogenic Bacteria and Fungi.
Ann. Bot. (Oxford, U. K.) 1946, 10 (39), 271-282.
28
ACS Paragon Plus Environment

Page 29 of 33
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10.
11.
12.
Prabhakaran, P.; Kale, S. S.; Puranik, V. G.; Rajamohanan, P. R.; Chetina, O.; Howard, J.
A. K.; Hofmann, H.-J.; Sanjayan, G. J. Sequence-Specific Unusual (1→2)-Type Helical
Turns in α/β-Hybrid Peptides. J. Am. Chem. Soc. 2008, 130 (52), 17743-17754.
Kheria, S.; Nair, R. V.; Kotmale, A. S.; Rajamohanan, P. R.; Sanjayan, G. J. The role of
N-terminal proline in stabilizing the Ant-Pro zipper motif. New J. Chem. 2015, 39 (5),
3327-3332.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Nair, R. V.; Kotmale, A. S.; Dhokale, S. A.; Gawade, R. L.; Puranik, V. G.;
Rajamohanan, P. R.; Sanjayan, G. J. Formation of a pseudo-β-hairpin motif utilizing the
Ant–Pro reverse turn: consequences of stereochemical reordering. Org. Biomol. Chem.
2014, 12 (5), 774-782.
13.
14.
15.
16.
17.
18.
Ramesh, V. V.; Priya, G.; Kotmale, A. S.; Gonnade, R. G.; Rajamohanan, P. R.;
Sanjayan, G. J. Multifaceted folding in a foldamer featuring highly cooperative folds.
Chem. Comm. 2012, 48 (91), 11205-11207.
Cameron, A. J.; Squire, C. J.; Edwards, P. J. B.; Harjes, E.; Sarojini, V. Crystal and NMR
Structures of a Peptidomimetic β-turn which provides facile synthesis of 13-membered
Cyclic Tetrapeptides. Chem. Asian J. 2017, (accepted) DOI: 10.1002/asia.201701422.
DeZoysa, G. H.; Sarojini, V. A Feasibility Study Exploring the Potential of Novel
Battacin Lipopeptides As Antimicrobial Coatings. ACS Appl. Mater. Interfaces 2017, 9
(2), 1373-1383.
De Zoysa, G. H.; Cameron, A. J.; Hegde, V. V.; Raghothama, S.; Sarojini, V.
Antimicrobial Peptides with Potential for Biofilm Eradication: Synthesis and Structure
Activity Relationship Studies of Battacin Peptides. J. Med. Chem. 2015, 58 (2), 625-639.
Qin, C.; Zhong, X.; Bu, X.; Ng, N. L. J.; Guo, Z. Dissociation of Antibacterial and
Hemolytic Activities of an Amphipathic Peptide Antibiotic. J. Med. Chem. 2003, 46 (23),
4830-4833.
Jin, L.; Bai, X.; Luan, N.; Yao, H.; Zhang, Z.; Liu, W.; Chen, Y.; Yan, X.; Rong, M.; Lai,
R. A Designed Tryptophan-and Lysine/Arginine-Rich Antimicrobial Peptide with
Therapeutic Potential for Clinical Antibiotic-Resistant Candida albicans Vaginitis. J.
Med. Chem. 2016, 59 (5), 1791-1799.
19.
Kohli, R. M.; Walsh, C. T.; Burkart, M. D. Biomimetic synthesis and optimization of
cyclic peptide antibiotics. Nature 2002, 418 (6898), 658-661.
29
ACS Paragon Plus Environment