Identification of the Ferric-Acinetobactin Outer Membrane Receptor in Aeromonas salmonicida subsp. salmonicida and Structure-Activity Relationships of Synthetic Acinetobactin Analogues

Article abstract of DOI:10.1021/acschembio.6b00805

Aeromonas salmonicida subsp. salmonicida, the causative agent of furunculosis in several fish species, produces acinetobactin and amonabactin as siderophores. In a previous study, we chemically characterized these siderophores and proposed a biosynthetic

Full text of DOI:10.1021/acschembio.6b00805

Subscriber access provided by UB + Fachbibliothek Chemie | (FU-Bibliothekssystem)
Article
Identification of the Ferri-Acinetobactin Outer Membrane Receptor
in Aeromonas salmonicida subsp salmonicida and Structure-
Activity Relationships of Synthetic Acinetobactin Analogues
Miguel Balado, Yuri Segade, Diego Rey, Carlos R Osorio,
Jaime Rodriguez, Manuel L. Lemos, and Carlos Jiménez
ACS Chem. Biol., Just Accepted Manuscript • DOI: 10.1021/acschembio.6b00805 • Publication Date (Web): 12 Dec 2016
Downloaded from http://pubs.acs.org on December 12, 2016
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.
ACS Chemical Biology 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 46
ACS Chemical Biology
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
Graphical abstract
38x31mm (300 x 300 DPI)
ACS Paragon Plus Environment

ACS Chemical Biology
Page 2 of 46
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
Identification of the Ferri-Acinetobactin Outer Membrane
Receptor in Aeromonas salmonicida subsp salmonicida and
Structure-Activity Relationships of Synthetic Acinetobactin
Analogues
Miguel Balado‡, Yuri Segade†, Diego Rey‡, Carlos R. Osorio,‡ Jaime Rodríguez†*,
Manuel L. Lemos,‡* and Carlos Jiménez†*
† Centro de Investigacións Científicas Avanzadas (CICA), Departamento de Química
Fundamental, Facultade de Ciencias, Universidade da Coruña, 15071 A Coruña,
Spain
‡. Department of Microbiology and Parasitology, Institute of Aquaculture,
Universidade de Santiago de Compostela, Campus Sur, Santiago de Compostela
15782, Spain
*Corresponding author, E-mail: manuel.lemos@usc.es; jaime.rodriguez@udc.es; carlos.jimenez@udc.es
KEYWORDS: Aeromonas, Aeromonas salmonicida, furunculosis, iron uptake,
siderophores, outer membrane siderophore receptors, acinetobactin, acinetobactin
analogues
1
ACS Paragon Plus Environment

Page 3 of 46
ACS Chemical Biology
1
2
3
4
ABSTRACT
5
6
7
8
Aeromonas salmonicida subsp. salmonicida, the causative agent of furunculosis in
several fish species, produces acinetobactin and amonabactin as siderophores. In a
previous study we chemically characterized these siderophores and proposed a
biosynthetic pathway based on genetic analysis. However, the internalization
mechanisms of ferric-acinetobactin and ferric-amonabactin remain largely unknown. In
the present study, we demonstrate that the outer membrane protein FstB is the ferric-
acinetobactin receptor in A. salmonicida since an fstB defective mutant is unable to grow
under iron limitation and does not use acinetobactin as an iron source. In order to study
the effect that structural changes in acinetobactin have on its siderophore activity, a
collection of acinetobactin-based analogues was synthesized, including its enantiomer
and four demethylated derivatives. The biological activity of these analogues on an
fstB(+) strain compared to an fstB(–) strain allowed structure-activity relationships to be
elucidated. We found a lack of enantiomer preference on the siderophore activity of
acinetobactin over A. salmonicida or on the molecular recognition by FstB protein
receptor. In addition, it was observed that A. salmonicida could not use acinetobactin
analogues when imidazole or a similar heterocyclic ring was absent from the structure.
Surprisingly, removal of the methyl group at the isoxazolidinone ring induced a higher
biological activity, thus suggesting alternative route(s) of entry into the cell that must be
further investigated. It is proposed that some of the synthetic acinetobactin analogues
described here could be used as starting points in the development of novel drugs against
A. salmonicida and probably against other acinetobactin producers like the human
pathogen Acinetobacter baumannii.
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
ACS Paragon Plus Environment

ACS Chemical Biology
Page 4 of 46
1
2
3
4
GRAPHICAL ABSTRACT
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
3
ACS Paragon Plus Environment

Page 5 of 46
ACS Chemical Biology
1
2
3
4
5
6
7
8
9
Iron is an essential trace nutrient for most organisms since it is a cofactor in key metabolic
processes.1 However, despite its high abundance in the Earth’s crust, iron bioavailability
is extremely low under aerobic conditions and at neutral pH because it is oxidized to
insoluble ferric hydroxides.2 In animal tissues iron is chelated by high-affinity iron-
binding proteins that restrict its availability to microbial pathogens.3 Therefore, most
pathogenic bacteria have developed sophisticated mechanisms to obtain iron from host
sources. These mechanisms are recognized as essential virulence factors for the survival
of microbes within their hosts. Siderophore synthesis is a widespread iron uptake
mechanism among bacterial pathogens. These compounds are low molecular weight, high
affinity iron chelating molecules that are secreted by bacteria to acquire iron from their
surroundings.4 Siderophores are present in most pathogens and it has been shown that
they are also important for the viability of marine bacteria in the natural environment.5
Aeromonas salmonicida subsp salmonicida (hereafter A. salmonicida) is a Gram-
negative Gamma- proteobacterium that causes furunculosis, a devastating disease that
affects salmonids and a variety of other non-salmonid fish, with a significant economic
impact in the aquaculture industry worldwide.6 In previous publications,7 we reported that
most A. salmonicida strains produce two catechol siderophores simultaneously:
acinetobactin (Acb, 1a), a siderophore first described in the human pathogen
Acinetobacter baumannii,8 and amonabactins such as amonabactin P750 (2), which were
previously found in Aeromonas hydrophila (Figure 1).9 While amonabactin genes are part
of a gene cluster conserved in most Aeromonas species, acinetobactin synthesis is
encoded by a second gene cluster only present in A. salmonicida. This cluster is
phylogenetically more related to the cluster that encodes the siderophore pseudomonine
in Pseudomonas entomophila10 than to the cluster encoding Acb (1a) in A. baumannii.11
Pseudomonine (3) and Acb (1a) only differ in the presence of the salicylic moiety in 3
rather than the catechol group in Acb (1a) (Figure 1). Most notably, the junction of the
amonabactin cluster, which provides 2,3-dihydroxybenzoic acid (DHBA), and the likely
horizontally acquired pseudomonine genes enable A. salmonicida to produce
simultaneously amonabactin and acinetobactin.7
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
4
ACS Paragon Plus Environment

ACS Chemical Biology
Page 6 of 46
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. Structures of natural siderophores (1-6) isolated from different bacteria.
Once siderophores are synthesized and secreted outside the bacterial cells, they bind Fe3+
ions to form a ferric-siderophore complex. The ferric-siderophore is then internalized by
the appropriate transport mechanism. In Gram-negative bacteria, ferric-siderophores are
transported across the outer and inner membranes by an energy-dependent system. The
components of such systems include a specific TonB-dependent ferric-siderophore
transporter (TBDT) located in the outer membrane, coupled to an ABC transporter that
catalyzes the final stages of ferric-siderophore transport through the cytoplasmic
membrane from the periplasm to the cytosol. In this regard, nothing is known about the
routes of entry of the ferric-siderophore complexes in A. salmonicida. It has been reported
that three putative TBDT are specifically induced in A. salmonicida under iron deficiency
and in vivo conditions.12 One of these was initially named FstA but has subsequently been
denoted as FstB in most recent studies due to the probable existence of two versions or
alleles. FstA/B is a putative siderophore receptor that is present in virulent strains and it
was used to develop PCR-based detection methods of A. salmonicida.13,14 Thus, although
fstA/B is a powerful molecular marker in the species, nothing is known about its biological
function.
In the last decade, new antimicrobial agents have been developed using the
‘Trojan horse’ approach, which involves the synthesis of a novel compound from an
5
ACS Paragon Plus Environment

Page 7 of 46
ACS Chemical Biology
1
2
3
4
5
6
7
8
9
antibiotic linked to a siderophore that uses a specific siderophore route of entry into the
cell.15 This approach seeks to avoid microbial resistance, thus enhancing sensitivity and
specificity.16 In order to develop rationally designed Acb-derived antimicrobials, it is
essential to know which TBDT serves as the entry route of ferric-Acb in A. salmonicida
and to define the moieties of the siderophore that play a role in the receptor recognition.
In the study reported here, we demonstrated that the outer membrane receptor
FstB is directly involved in the uptake of Acb (1a) in A. salmonicida. Furthermore,
several synthetic analogues of Acb (1a) were prepared, including its enantiomer, and their
capacity to serve as iron sources to A. salmonicida was evaluated. These analyses allowed
us to establish very valuable structure-activity relationships that must be taken into
consideration for the possible preparation of Acb-antibiotic conjugates.
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
6
ACS Paragon Plus Environment

ACS Chemical Biology
Page 8 of 46
1
2
3
4
RESULTS AND DISCUSSION
5
6
7
8
9
Sequence analysis of the acinetobactin genes of A. salmonicida revealed the existence of
two candidate gene clusters that potentially encode ferric-siderophore uptake systems
(Figure 2).7,17 The first gene cluster could be related to a hydroxamate-type siderophore
transport, being the closest homologue genes conserved in several bacterial genomes and
annotated as fhuABCDE. Interestingly, FhuA shows homology to the putative
ferrichrome receptor of Aeromonas hydrophila (FhuA, ac. Nº 4487379; 92% identity).
The second cluster includes the uncharacterized fstB (encoding a TBDT) and genes for
components of an ABC transporter (asuBECDGF). These show significant homology
(between 43 to 79% aa identity) with the TBDT and ABC transporters involved in
pseudomonine transport in P. entomophila10 and acinetobactin uptake in A. baumannii.11
Although FstB is homologous to the ferric pseudomonine receptor of P. entomophila
(GenBank accession No CAK15297) and to the ferric acinetobactin receptor of A.
baumannii (GenBank accession No YP_001713025), it only shares with those receptors
46 and 43% aa identity, respectively. Therefore, although a possible role of FhuA in
ferric-siderophore acquisition cannot be ruled out, we hypothesize that fstB could encode
the ferric-Acb TonB-dependent outer membrane receptor, which is indispensable for the
ferric-siderophore internalization.
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 2. Scheme of the acinetobactin biosynthesis and transport gene cluster.7
fstB is essential for growth under iron starvation of A. salmonicida strains lacking
amonabactin production
In order to assess the role of fstB and fhuA in ferric-acinetobactin transport, individual in-
frame deletions were constructed by allelic exchange in strain RSP74.1. A. salmonicida
RSP74.1 is a virulent strain that was previously used to characterize the asb genes.18 It is
a natural producer of Acb (1a) but it does not produce amonabactins (i.e. compound 2)
7
ACS Paragon Plus Environment

Page 9 of 46
ACS Chemical Biology
1
2
due to a deletion in the amoG gene.7
3
4
5
6
7
8
When individual mutants were cultured under iron-sufficient conditions (CM9
plus 10 μM ferric sulfate), significant differences in growth levels were not observed with
respect to the parental strain (Figure 3). However, under iron-restricted conditions (CM9
plus 75 µM 2,2'-dipyridyl), the ∆fstB mutant was severely affected in its ability to grow
under iron starvation and it showed an equivalent phenotype to that of the mutant that is
unable to produce Acb (1a) (RSP74.1∆asbD). An analysis of the culture supernatants by
the CAS assay showed similar levels of siderophore production in the ∆fstB mutant with
respect to the parental strain (Figure 3). At an OD600 of approximately 1 in CM9 plus 50
µM 2,2'-dipyridyl, the parental strain and the mutant showed CAS values of ca. –0.38
(lower values indicate higher concentrations of siderophore).19 This demonstrates that the
growth phenotype of the ∆fstB mutant is not due to the lack of siderophore production. It
is noteworthy that the RSP74.1∆fhuA mutant is not affected in its ability to grow under
iron-limiting conditions nor in its siderophore production (Figure 3). Complementation
of the fstB mutation, by providing in trans an intact copy of the fstB gene, restored growth
under iron limitation at the same level of wild type strain (Figure 3). Altogether, these
results clearly indicate that the TBDT FstB is essential for RSP74.1 growth under iron
deprivation and it must be directly involved in the utilization of Acb (1a) by A.
salmonicida.
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. Growth (OD600) after 12 h of incubation of A. salmonicida subsp salmonicida
(RSP74.1 wild type, RSP74.1ΔasbD, RSP74.1ΔfstB, RSP74.1ΔfhuA and RSP74.1 ΔfstB
8
ACS Paragon Plus Environment

ACS Chemical Biology
Page 10 of 46
1
2
3
4
5
6
7
8
complemented with a wild type copy of the gene fstB) strains in CM9 minimal medium
supplemented with Fe2(SO4)3 (10 μM) or with addition of the iron chelator 2,2'-dipyridyl
(75 μM).
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
fstB is the ferric-acinetobactin receptor in A. salmonicida
In order to demonstrate that FstB is the receptor for Acb (1a) in A. salmonicida, we first
performed cross-feeding bioassays. A. salmonicida mutants RSP74.1ΔasbD and
RSP74.1ΔfstB were used as indicator strains and A. salmonicida wild type strains
RSP74.1 (produces Acb (1a) only) and VT45.1 (produces Acb (1a) and amonabactins
(i.e. compound 217) and Acinetobacter baumannii strain ATCC19606 (produces Acb
(1a)) as test strains (Table S1). The results showed that all of the A. salmonicida and A.
baumannii acinetobactin producer strains could promote growth of the ∆asbD mutant but
not of the ∆fstB mutant (Table 1, Figure S1). However, A. salmonicida strains VT45.1
and VT45.1∆asbD (a mutant that does not produce Acb (1a)) could promote growth of
both mutants (Table 1, Figure S1) since they produce amonabactins, which can be utilized
through its own receptor by all mutants, with its transport being independent of FstB.
To confirm the above results, we also tested the utilization of purified
amonabactin P750 (2) and Acb (1a) by the indicator mutants. Amonabactin P750 (2) was
used as an iron source by both mutants but the ∆fstB mutant was unable to utilize natural
Acb (1a), although it could still use amonabactin P750 (2) (Table 1, Figure S1). As a
comparison, a ∆fhuA mutant used Acb (1a) as efficiently as the parental strain (results
not shown). These results reinforced the idea that fstB encodes the acinetobactin TBDT
in A. salmonicida and they demonstrate that FstB is directly involved in the uptake of
ferric Acb (1a).
9
ACS Paragon Plus Environment

Page 11 of 46
ACS Chemical Biology
1
2
3
4
Table 1. Results of cross-feeding experiments with two indicator strains.
5
6
7
8
9
Ability to cross-feed
RSP74.1ΔfstB
RSP74.1ΔasbD
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
Strains
Aeromonas salmonicida subsp salmonicida
RSP74.1 a
–
–
–
–
+
+
+
–
+
+
+
+
RSP74.1ΔasbD b
RSP74.1ΔfstB a
RSP74.1ΔfhuA a
VT45.1a,c
VT45.1ΔasbD c
Acinetobacter baumannii
ATCC 19606 a
–
+
Pure natural siderophores
Acb (1a)
–
+
+
+
Amonabactin P750 (2)
Acinetobactin producer
a
b
Does not produce any siderophore
c
Amonabactin producer
fstB is the most abundant allele of the acinetobactin receptor gene in A. salmonicida
The first sequence of a ferric-siderophore transporter in A. salmonicida subsp.
salmonicida was deposited in GenBank in 1995 and was named fstA (GI:1143382). Since
then other versions (alleles) of this gene, which were called either fstA or fstB, have been
sequenced and deposited in GenBank. The two alleles, fstA and fstB, have a nucleotide
identity of 98% (Supporting Information, Figure S2); however, the protein similarity is
only 86% due to the existence of certain regions that do not align to each other
(Supporting Information, Figure S3). FstA and FstB show the characteristic TBDT
domain architecture composed of a 22-stranded transmembrane β-barrel that encloses a
globular plug domain. The N- and C-terminals are conserved among TBDTs so they are
necessary for TonB interaction (N-terminal) and for the correct protein assembly in the
outer membrane (C-terminal). In addition, the ligand binding site is usually located in the
extracellular side of the plug domain.20 The unmatched regions between FstA and FstB
are located outside the most probable regions related to protein functionality. The
differences in the amino acid sequences are mostly located in the extracellular loops and
10
ACS Paragon Plus Environment

ACS Chemical Biology
Page 12 of 46
1
2
3
4
5
6
7
8
outside the plug-domain and C- and N-terminals, so they should not affect protein
function, although this possibility cannot be completely ruled out and must be further
investigated.
Currently the distribution of fstA/B among A. salmonicida strains is unknown. A
nested PCR test was used to detect fstA/B and differentiate between fstA or fstB alleles in
a collection of 37 A. salmonicida strains isolated from diseased freshwater and marine
fish. Interestingly, all A. salmonicida strains tested harbour the fstB version (results not
shown). This result is consistent with the fact that only one of the 18 different fstA/B
sequences deposited in GenBank corresponds to fstA.
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
Synthesis of Acb analogues (1b and 7–10)
The isolation and structure elucidation of Acb (1a), as the siderophore produced by A.
baumannii ATCC19606, was reported in 1994.8 However, it was suggested in 2009 that
the reported structure 4 was an unstable intermediate, which was later named as pre-
acinetobactin (Pre-Acb, 4), and the correct structure was proposed as 1a, which
corresponds to an intramolecular nucleophilic substitution from 4 to 1a.21 A complete
study about the acinetobactin isomerization at different pH was recently published.22 The
same type of rearrangement was previously observed in the natural analogue
pseudomonine (3),23 a blue-fluorescent siderophore, which was first isolated from
Pseudomonas fluorescens AH2 obtained from Nile perch.24 The structure of Acb as 1a,
including its absolute configuration, was confirmed by chemical synthesis.25
With the aim of elucidating some structure-activity relationships (SARs) in Acb (1a),
analogues ent-acinetobactin (ent-Acb, 1b) and four demethylacinetobactin derivatives
(7–10) were prepared (Figure 4). The structural design of these derivatives was based on
its simplicity in order to facilitate the preparation of conjugates to be used as chemical
probes. Information from SARs could be very useful for the development of new
antimicrobial strategies based on iron uptake mechanisms.
11
ACS Paragon Plus Environment

Page 13 of 46
ACS Chemical Biology
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 4. Structures of the synthetic acinetobactin analogues 1b and 7-10.
a) Synthesis of ent-Acb (1b)
In an effort to evaluate the influence of the stereochemistry of the chiral centres on the
siderophore activity, we designed the synthesis of ent-Acb (1b) as shown in Scheme 1
based on the previous synthesis of Acb (1a) developed by Takeuchi et al.25 .
12
ACS Paragon Plus Environment

ACS Chemical Biology
Page 14 of 46
1
2
Scheme 1. Synthesis of ent-Acb (1b) a
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
The synthesis was initiated with the access of a key amide 14, which is easily
obtained by the coupling of 2,3-diisopropyloxybenzoic acid (12)26 with benzyl L-
threonine (13).27 The optimized conditions for this coupling to give the highest yield
(64%) involved the use of 2.5 equivalents of Et3N and TBTU as the coupling agent
(instead of EDC or EDC/HOBt).28
13
ACS Paragon Plus Environment

Page 15 of 46
ACS Chemical Biology
1
2
3
4
5
6
7
8
9
Dehydrative-cyclization of amide 14 mediated by SOCl2 in CH2Cl2 at room
temperature gave the corresponding oxazoline 15, which was debenzylated by
hydrogenolysis to afford the carboxylic acid 16 in a very good yield in both steps (Scheme
1). Dehydrative-cyclization led to an inversion of configuration of the oxymethine carbon
(C-9) in 15, as supported by NOE cross peaks in 16 between protons H-8 and H-9 and
confirming the cis disposition of these protons (see Supporting Information).
In parallel, the commercially available histamine dihydrochloride (17) was
transformed into O-benzyloxyhistamine (20) via N-tert-butoxycarbonyl-O-
benzyloxyamine (19) following the methodology described by Takeuchi et al.25 (Scheme
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
With these products in hand, oxazolinic acid 16 was coupled with O-
benzyloxyhistamine (20) using EDC·HCl and HOBt to afford 21 in 43% yield. The yield
could be improved to 69% when the reaction was performed with CDI in the absence of
base. This coupling afforded the sterically stable trans-isomer 21 rather than the cis-
isomer, probably due to epimerization of the activated acyl intermediate. The NOE cross
peak between methyl group protons at C-9 and H-8 along with the lack of such correlation
between protons H-8 and H-9 in the NOESY spectrum of 21 is consistent with the
proposed trans-configuration (see the Supporting Information). A similar epimerization
was also reported in the synthesis of another Acb analogue.25 Debenzylation of 21
afforded the corresponding oxazoline 22 in very good yield and this was heated under
reflux in MeOH to give the expected rearrangement of oxazoline 22 to isoxazolidinone
23. This conversion was first observed from pre-pseudomonine (5) to pseudomonine (4)23
and it confirms the electrophilic character of the oxazoline ring in 22, which is
intramolecularly attacked by the nucleophilic hydroxamate oxygen to give the
isoxazolidinone ring in 23. However this rearrangement does not occur in the case of the
siderophore anguibactin (6) (Figure 1), which bears a thiazoline ring in 6 instead an
oxazole ring in 3 and 5, probably due to the decreased electrophilicity of Cβ of the
thiazoline ring relative to that of an oxazoline ring.22
Finally, removal of the isopropyl groups with BCl3 in dry CH2Cl2 at low
temperature afforded, after HPLC purification, ent-Acb (1b·TFA) in 23% global yield in
9 steps from 2,3-dihydroxy benzoic acid (Scheme 1).
Ent-Acb (1b.TFA) has the same NMR data and the same HPLC retention time as
the natural TFA salt isolated from A. salmonicida subsp salmonicida Acb (1a.TFA). The
absolute configuration of ent-Acb (1b.TFA) was confirmed by its circular dichroism (CD)
14
ACS Paragon Plus Environment

ACS Chemical Biology
Page 16 of 46
1
2
3
4
5
6
7
8
spectrum, which showed opposite Cotton effects, and a similar optical rotation, but with
25
opposite signs, when compared Acb (1a.TFA) ([α]D25 = +22 for 1b.TFA instead of [α]D
= –29.7 for 1a.TFA) (see the Supporting Information).
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
b) Synthesis of demethylacinetobactin analogues (7–10)
In order to compare the influence of the methyl group on the isoxazolidinone ring and the
presence of similar heterocyclic rings such triazoles instead of imidazole on the
siderophore activity of acinetobactin analogues, four demethylacinetobactin derivatives
(compounds 7–10) were prepared from a common intermediate 26 (Scheme 2).
Compound 26 was synthesized from 2,3-dihydroxybenzoic acid (11), which was
converted to 2429 and then coupled to the commercially available D-cycloserine (25) using
an improved procedure employed by Walsh et al.21
The first demethyl Acb, compound 7, was designed due to its simplicity and
because it contains a primary amine group that may provide a possible anchor point for
conjugation. Thus, the synthesis of 7 began with the N-alkylation of intermediate 26 with
mesylate 27, which was prepared in a straightforward manner from the commercially
available 3-benzyloxycarbonylaminopropanol and methanesulfonyl chloride, to give
compound 28. The subsequent removal of Cbz and Bn protecting groups in 28 by
hydrogenolysis afforded, after HPLC purification, 7·TFA (Scheme 2).
15
ACS Paragon Plus Environment

Page 17 of 46
ACS Chemical Biology
1
2
Scheme 2. Synthesis of analogues 7 and 8a
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
The next demethylacinetobactin, compound 8, was prepared following a procedure
similar to that reported by Walsh et al.21 N-Alkylation of the intermediate 26 with the
substituted imidazole 18, prepared previously for the synthesis of ent-Acb (1b), gave 29,
which was deprotected by hydrogenolysis and purified by HPLC to afford analogue
8·TFA (Scheme 2).
A ‘click-chemistry’ reaction involving the efficient Cu(I)-catalyzed azide-alkyne
couplings was used for the synthesis of the triazole analogues 9 and 10 (Scheme 3). This
reaction is an extremely useful tool for preparation of conjugates.30 Mesylate 30, prepared
from commercially available homopropargyl alcohol, was coupled with intermediate 26
using Na2CO3 and NaI to give compound 31. Subsequent coupling of alkyne 31 with
trimethysilyl azide in the presence of Cu(I)31 furnished 32 in quantitative yield.
16
ACS Paragon Plus Environment

ACS Chemical Biology
Page 18 of 46
1
2
3
4
5
6
Subsequent debenzylation of 32 followed by purification by HPLC afforded the triazole
analogue 9·TFA (Scheme 3).
7
8
9
Scheme 3. Synthesis of analogues 9 and 10a
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
On the other hand, azide derivative 33, prepared by displacement of the mesylate
group in 30 with sodium azide, was coupled with alkyne 31 in a copper(I)-catalyzed 1,3-
dipolar cycloaddition reaction following the procedure reported by Meldal et al.32 In this
way, the 1,4-substituted triazole 34 was obtained as the only regioisomer in quantitative
yield. The regioselectively of the reaction was confirmed by the long range proton-carbon
cross peak between the H-15 methylene protons at 4.32 ppm and the C-14 aromatic CH
carbon at 126.1 ppm in the HMBC experiment on compound 34 (see Supporting
Information). Debenzylation of the resulting product followed by HPLC purification
afforded the desired triazole analogue 10·TFA (Scheme 3).
17
ACS Paragon Plus Environment

Page 19 of 46
ACS Chemical Biology
1
2
Scheme 4. Alternative synthesis of intermediates 32 and 34 a
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
Alternatively, the triazole analogues 9 and 10 could be obtained by reversing the
order of the reactions by preparing intermediates 35 and 36 as shown in Scheme 4. The
number of reaction steps in both strategies is the same but the global yield was slightly
lower on using the latter approach.
Biological evaluation of ent-Acb (1b) and demethyl-Acb analogues 7–10
In previous sections it was shown that fstB encodes the ferric-acinetobactin receptor in A.
salmonicida. Moreover, several acinetobactin analogues were synthesized with the aim
of studying the interaction between acinetobactin and its cognate receptor FstB. All these
analogues showed CAS values indistinguishable from those of acinetobactin, which
indicates that all them bind iron efficiently. In this section, the siderophore activity of all
analogues was tested by growth promotion of an indicator strain fstB(+) in relation to an
fstB(–) strain under iron-deficient conditions at physiological pH of 7.0.
a) Biological activity of ent-Acb (1b)
The results of growth promotion assays on natural Acb (1a) and its synthetic enantiomer
ent-Acb (1b) against both fstB(+) and fstB(–) strains are shown in Figure 5 and Figure S4.
The results show that ent-Acb (1b) promoted growth under iron-deficient conditions of
the fstB(+) strain at the same level as Acb (1a) and in a dose-dependent manner (Figure
6). However, neither of these two compounds, Acb or ent-Acb, could promote the growth
of the fstB(–) strain (Figure 5, Figure S4). This observation demonstrates not only that
ent-Acb (1b) has siderophore activity but also that the FstB receptor (acinetobactin
receptor) is the route of entry into the cell, thus confirming the results described above.
18
ACS Paragon Plus Environment

ACS Chemical Biology
Page 20 of 46
1
2
3
4
5
6
7
8
9
This observation reveals the lack of preference of the Acb enantiomer on the siderophore
activity in A. salmonicida or on the molecular recognition by the FstB protein receptor.
We previously found a similar lack of stereoisomeric influence of the amino acid scaffold
on the siderophore activity in a series of analogues of vanchrobactin,33 a siderophore
isolated from Vibrio anguillarum.34 These observations on Acb (1a) and vanchrobactin
contrast those reported for enterobactin and pyochelin in relation to their corresponding
enantiomers. Ent-enterobactin does not promote growth of E. coli K12 under iron-
deficient conditions. Although ent-enterobactin binds to the outer membrane receptor
FepA with similar affinity as enterobactin, however at a later stage, the Fes esterase,
responsible for degrading the molecule, does not accept the ent-enterobactin or its iron
complex as a substrate.35 On the other hand, iron uptake and transcriptional regulation
with pyochelin and ent-pyochelin are stereospecific processes in Pseudomonas
aeruginosa and P. fluorescens, respectively. That stereospecificity is ensured not only by
the outer membrane siderophore receptors but also by the cytosolic PchR regulators.36
Indeed, the results obtained here help to expand our knowledge of the role of chirality in
iron delivery.
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 5. Biological activity of ent-Acb (1b) (blue lines) compared to Acb (1a) (brown
lines). The graphs show the growth achieved after 12 h incubation in CM9 with different
19
ACS Paragon Plus Environment

Page 21 of 46
ACS Chemical Biology
1
2
3
4
5
6
7
8
9
concentrations of 2,2'-dipyridyl (from 30 to 70 µM). Indicator strains were an fstB(+)
strain (RSP74.1∆asbD) that is able to use Acb (1b) (Continuous lines) and an fstB(–)
strain (RS74.1∆asbD∆fstB) that is unable to use Acb (1b) (dashed lines). The grey shaded
area represents basal growth without addition of any siderophore.
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. Growth achieved by A. salmonicida fstB(+) strain in CM9 medium with 40 µM
dipyridyl supplemented with a 0.01, 0.1 and 1 µg of Acb (1a), ent-Acb (1b) or one of the
analogues 7-10.
b) Biological evaluation of demethylacinetobactin analogues 7–10
The biological activity of synthetic demethylacinetobactin analogues 7–10 was compared
to the activity of natural Acb (1a) using growth promotion assays against fstB(+)
(RSP74.1ΔasbD) and fstB(–) (RSP74.1ΔasbDΔfstB) strains (Figure 7). When 1 µg of
natural Acb (1a) was added to the fstB(+) strain cultures the maximum growth achieved
was 1.4 (OD600) at 40 µM 2,2'-dipyridyl, while under the same conditions the fstB(–)
strain reached a maximum OD of only 0.55. This finding can be easily explained by the
fact that this strain is unable to use Acb (1a) (see above).
The demethylacinetobactin analogues, compound 7 (bearing a propyl amino
moiety instead of ethylimidazole group) and compound 10 (carrying a substituted
20
ACS Paragon Plus Environment

ACS Chemical Biology
Page 22 of 46
1
2
3
4
5
6
7
8
9
ethyltriazole ring), were almost unable to promote the growth of any of the indicator
strains and hence they lack biological activity against both fstB(+) and fstB(–) strains
(Figure 7, Figure S4). In contrast, the demethylacinetobactin analogues, compound 8
(bearing an imidazole ring) and compound 9 (bearing a triazole ring), were clearly able
to promote the growth of the fstB(+) strain to a significantly higher extent than the natural
Acb (1a), with a maximum OD of 1.6 (Figure 7, Figure S4) and a two-fold growth
increase at 50 µM 2,2'-dipyridyl with respect to Acb (1a). Both compounds (8 and 9)
promoted the growth of A. salmonicida in a dose-dependent manner (Figure 6). The use
of these compounds (8 and 9) by the fstB(–) strain was analyzed and it was observed that,
although their biological activity was lower in relation to the fstB(+) strain, they retained
some growth promotion activity on the fstB(–) strain compared with Acb (1a). Compound
8 retained ca. 75% of the activity against the fstB(–) strain when compared to fstB(+)
(OD600 = 1.2 vs. 1.5 at 45 µM 2,2'-dipyridyl) (Figure 7). This result suggests that
compound 8 could mainly use a transporter alternative to the FstB receptor to enter into
the cell. On the other hand, compound 9 also showed growth-promotion activity on the
fstB(+) strain at about the same level as 8, but the growth promotion on the fstB(–) strain
was ca. 50% in relation to fstB(+) (OD600 = 0.7 vs. 1.5 at 45 µM 2,2'-dipyridyl) (Figure
7). This finding suggests that compound 9 must use both the FstB receptor and an
alternative route of entry to be internalized. Further work is needed to identify this
alternative transporter.
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
Additionally, the possibility that FhuA could participate in the uptake of
compounds 8 and/or 9 was assessed by measuring the biological activity of these
compounds in a strain that lacks FstB and FhuA (RSP74.1∆asbD∆fstB∆fhuA) compared
to an fstB(–) strain with an intact FhuA (RSP74.1∆asbD∆fstB). The results showed
identical phenotypes for the two strains (data not shown) and this rules out FhuA as an
alternative route of entry for compounds 8 and 9.
In summary, all of the results discussed above suggest that the presence of an
imidazole (1a and 8), or a similar ring such as a triazole without substitution (9), is
required to promote the growth of the fstB(+) indicator strain. Furthermore, the presence
of a primary amino group (7) or a propylamino substituted triazole (10) makes the
compound unable to promote growth since it does not serve as an iron source for A.
salmonicida. The findings suggest that the recognition of acinetobactin by its cognate
receptor FstB mostly resides in the presence of imidazole or a similar heterocyclic ring in
its structure. Furthermore, the absence of the methyl group on the isoxazolidinone ring,
21
ACS Paragon Plus Environment

Page 23 of 46
ACS Chemical Biology
1
2
3
4
5
6
7
8
9
in combination with the presence of either an imidazole or a triazole ring (8 and 9), seems
to open alternative route(s) of entry into the cell, which significantly increased the
biological activity. This finding could be useful for the rational design of novel
acinetobactin-based 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
Figure 7. Biological activity results for Acb 1b and analogues 7–10 (blue lines) compared
to Acb (1a) (brown lines). The graphs show the growth achieved after 18 h of incubation
in CM9 with different concentrations of 2,2'-dipyridyl (from 30 to 70 µM) with the
addition of acinetobactin analogue (blue lines) compared to Acb (1a, brown lines). The
indicator strains were an fstB(+) strain (RSP74.1∆asbD) that is able to use Acb (1a)
(continuous lines) and an fstB(–) strain (RS74.1∆asbD∆fstB) that is unable to use Acb
(1a) (dashed lines). The grey shaded area represents basal growth without addition of any
compound.
22
ACS Paragon Plus Environment

ACS Chemical Biology
Page 24 of 46
1
2
3
4
5
6
7
8
9
CONCLUSIONS
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
An understanding of the correlations between the structure and the chemical and
biological activity of siderophores is crucial for the development of novel applications of
the bacterial iron uptake mechanisms based on siderophores.37 This information is very
important, for instance, for the preparation of conjugates to be used as biosensors to study
these mechanisms,38 for the development of novel therapies against iron-overload related
disease,39 for the preparation of new antimicrobial agents that target bacterial pathways
for the acquisition of iron, for a Trojan horse strategy to develop modified molecules with
antimicrobial activity,40 or for the synthesis of bioconjugates of siderophores to be tested
as antibacterial vaccines.41 In a recent study, we demonstrated that all virulent strains of
the fish pathogen A. salmonicida produce acinetobactin as a siderophore.7
In the present study, by means of mutant construction, we identified the TonB-dependent
outer membrane receptor necessary for the internalization of ferric-Acb and demonstrated
that the fstB gene encodes this receptor. In order to establish some structure-activity
relationships, several isoxazolidinone analogues of Acb were synthesized, including its
enantiomer (ent- Acb), and their biological activity was evaluated by growth promotion
assays on fstB(+) and fstB(–) strains. The observed lack of enantiomer preference on the
recognition of Acb by FstB is worth highlighting. In addition, it seems that A. salmonicida
cannot use Acb isoxazolidinone analogues if imidazole or a similar heterocyclic ring is
not present in the structure. Surprisingly, removal of the methyl group at the
isoxazolidinone ring leads to higher levels of biological activity, which suggests
alternative route(s) of entry into the cell. More specifically, we suggest that compound 9
is a potential candidate as a vector in a Trojan horse strategy to develop new treatments
against furunculosis in fish. Moreover, the fact that acinetobactin is also the main
siderophore used by Acinetobacter baumannii opens the door to the development of novel
drugs against this relevant human pathogen for which multi-drug-resistant (MDR)
clinical isolates are frequently found.42
23
ACS Paragon Plus Environment

Page 25 of 46
ACS Chemical Biology
1
2
3
4
5
6
METHODS
A. Biological methods.
Bacterial Strains, Plasmids and Media. Strains and plasmids used, as well as
those derived from this study are listed in Supplementary Information (Table S1). A.
salmonicida subsp. salmonicida strains were grown at 25 ºC in Tryptic Soy Agar and
Broth (Pronadisa) supplemented with 1% NaCl (TSA–1 and TSB–1, respectively), as
well as in M9 minimal medium43 (at physiological pH of 7.0) supplemented with 0.2%
Casamino Acids (Difco) (CM9). Escherichia coli strains were routinely grown at 37 ºC
in Luria Bertani (LB) medium (Pronadisa) or LB supplemented with the appropriate
antibiotics. Ampicillin sodium salt (Ap) (Sigma–Aldrich) was used at 50 µg mL-1;
tetracycline (Tc), at 12 µg mL-1 and gentamicin (Gm) at 15 µg mL-1 (final concentrations).
All stocks were filter sterilized and stored at –20ºC. The iron chelator 2,2´–dipyridyl
(TCI) was dissolved in distilled water to prepare a stock solution at 20 mM that was added
to the sterile media at appropriated concentrations.
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
DNA Manipulations and Bioinformatics Tools. Total genomic DNA from A.
salmonicida was purified with the Easy–DNA kit (Invitrogen). Plasmid DNA purification
and extraction of DNA from agarose gels were carried out using kits from Fermentas
(Thermo–Fisher). PCR reactions were routinely carried out in a T–Gradient Thermal
Cycler (Biometra), with Taq polymerase BioTaq (Bioline). FstA/B alignment were done
with BLAST program.44 SignalP 4.1 server was used to predict the presence and location
of signal peptide cleavage site.45 Domain content and prediction of the protein structures
were done using Pfam protein families database46 and SWISS-MODEL Server.47 For
FstB 3D structure modelling the crystal structure of Pyoverdine outer membrane receptor
FpvA from Pseudomonas aeruginosa was used as template (SMTL id: 1xkh.1.).48
Detection of Acinetobactin Receptor fstA or fstB version. A nested PCR was
designed to detect the suitable version of the acinetobactin receptor (fstA or fstB). External
primer asbE_F and asbE_R were used to amplify fstA/B gene resulting in an amplicon of
1975 bp. One microliter of a 1/100 dilution in water of the first-round product was used
as template for the second round using primers fstB_F and fstB_R that are specific of fstB
version. Both PCR were done in a total volumen of 25 µl with Taq polymerase BioTaq
(Bioline) according manufacture instructions. The cycling conditions were 95°C for 180
s, followed by 30 cycles of 94°C for 45 s, 57°C for 45 s, and 72°C for 90 s.
24
ACS Paragon Plus Environment

ACS Chemical Biology
Page 26 of 46
1
2
3
4
5
6
7
8
9
Construction of fstB and fhuA Mutants. In-frame deletions of fstB and fhuA in A.
salmonicida subsp. salmonicida RSP74.1 were constructed by allelic exchange as
previously described.18 PCR amplifications of two fragments of each gene and flanking
regions, when ligated together result in an in–frame (nonpolar) deletion. Primers used are
listed in Supporting Information (Table S2). Each deleted gene construction was ligated
into the suicide vector pKEK229.49 The resulting plasmids were mated from E. coli S17-
1-λpir50 into A. salmonicida subsp. salmonicida RSP74.1 wild type strain (TcR) and into
previously constructed RSP74.1ΔasbD mutant strains,18 and exconjugants with the
plasmid (ApR) integrated in the chromosome by homologous recombination were
selected. A second recombination event was obtained by selecting for sucrose (15%)
resistance and further checking for plasmid loss and for allelic exchange. This process
led to the generation of A. salmonicida subsp. salmonicida mutants RSP74.1ΔfhuA,
RSP74.1ΔfstB, RSP74.1ΔasbDΔfstB and RSP74.1ΔasbDΔfhuA. Deletion of the parental
gene was checked by DNA sequencing of the region involved to ensure that mutations
were in-frame. For fstB gene complementation the ORF was PCR-amplified (primers
listed in Supporting Information Table S3) with Hi-Fidelity Kapa Taq (Kapa), cloned into
pHRP309 vector and mobilized from E. coli S17-1 ʎ-pir into the A. salmonicida R
P74.1ΔfstB mutant.
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
Growth under Iron Limiting Conditions and Test of Siderophore Production.
To test the ability of A. salmonicida subsp. salmonicida deleted mutants to grow under
iron limiting conditions, overnight cultures in LB of the parental and mutant strains were
adjusted to an optical density (OD600) of 0.5 and diluted 1:100 in CM9 minimal medium
containing the iron chelator 2,2´-dipyridyl at 75 µM. Cultures were incubated at 25 ºC
with shaking at 150 rpm, and growth (OD600) was measured after 12 h incubation.
Siderophore production was measured using the chrome azurol-S (CAS) liquid assay,19
which detects the presence of iron-chelating siderophore molecules. For siderophore
production, strains were grown at 50 µM 2,2´-dipyridyl to allow enough growth to make
siderophore secretion detectable. A non-inoculated CM9 sample containing 2,2´-
dipyridyl at appropriate concentration and a sample from the asbD mutant were used
respectively as spectrophotometric blank and as negative control for CAS liquid assay.
Growth curves and CAS assays were carried out in triplicate, and results shown are the
means of three independent experiments.
25
ACS Paragon Plus Environment

Page 27 of 46
ACS Chemical Biology
1
2
3
4
5
6
7
8
9
Cross-feeding Assays. Bioassays were designed to determine the role of the
TBDTs FstB and FhuA in uptake of Acb (1a). Mutant strains RSP74.1∆asbD,
RSP74.1∆fhuA and RSP74.1∆fstB were used as indicator strains. They were inoculated
into CM9 minimal medium containing the iron chelator 2,2´-dipyridyl at 150 µM, a
concentration higher than the MIC for the wild type strain. A. salmonicida wild type
strains RSP74.1 (produces only Acb (1a)) and VT45.1 (produces Acb (1a) and
amonabactin) and Acinetobacter baumannii strain ATCC 19606 (produces Acb (1a))
were cultured on LB agar plates, cells were harvested with a sterile loop and placed on
top of the indicator strain plates. Purified Acb (1a) and amonabactin P750 (2a), were
deposited on the plates surface in paper discs (6 mm diameter) containing 25 µg of each
siderophore. The results were scored as positive when a growth halo of the indicator
strains was visible around discs. A disc containing 10 µl of a solution of Fe2(SO4)3 10 µM
was used as positive growth control.
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
Quantification of the Biological Activity of ent-Acinetobactin and Compounds
7-10. The potential of ent-Acb (1b) and Acb analogues (compounds 7-10) to promote
the growth of A. salmonicida was compared to natural Acb (1a). As indicators of the
biological activity, we used a fstB(+) strain and a fstB(-) strain that were cultured under
increasing iron-deficient conditions (concentration of 2,2’-dipyridyl from 30 to 70 µM).
As fstB(+), we used the mutant strain RSP74.1∆asbD that is able to efficiently acquire
acinetobactin although cannot produce it (Figure 7). As fstB(-) strain, we used the double
mutant RSP74.1∆asbD∆fstB that is unable to produce or transport acinetobactin. Both
fstB(+) and fstB(-) strains showed identical MIC of 2,2’-dipyridyl (ca. 35 µM), which is
significantly lower than that of RSP74.1 wild type (around 100 µM). The addition to the
culture medium of an Acb analogue that can be used as iron source by these indicator
strains will promote their growth and will increase the MIC of 2,2’-dipyridyl.
Consequently, we assumed that the maximum growth achieved (OD600) by each strain
will be related to the biological activity of the compounds tested.
Mid-log-phase A. salmonicida cultures of both indicators (fstB(+) and fstB(-)) in
TSB-1 were adjusted to an OD600=0.5 and diluted 1/20 in CM9 (at physiological pH of
7.0) containing increasing inhibitory concentrations of 2,2’-dipyridyl, from 30 and 70
µM. Growth promotion assays were performed using 96-well microtiter plates. Each well
contained 100 µl of CM9, supplemented with the appropriated concentration of 2,2’-
dipyridyl, inoculated with the indicator strain. One µg of the compound to be tested, either
26
ACS Paragon Plus Environment

ACS Chemical Biology
Page 28 of 46
1
2
3
4
5
6
7
8
9
natural Acb (1a), ent-Acb (1b) or some of the Acb analogues 7-10, was added to each
plate well, giving final concentrations of 2.9 µM for Acb and ent-Acb; 3.0 µM for
compounds 8 and 9; 3.4 µM for compound 7 and 2.6 µM for compound 10. The plates
were incubated for 18 h at 25 ºC under agitation at 150 rpm, and the growth achieved was
evaluated by measuring the OD600 with an Infinite 200 PRO (TECAN) microplate reader.
In the case of ent-Acb (1b) it was measured after 12 h. Each sample was duplicated within
each plate and the assays were repeated three times. All the replicas showed similar
results. Negative controls were made with the no addition of Acb (1a) (or any of the
compound tested), and a plate well with non-inoculated CM9 was used as
spectrophotometric blank. Mean and standard deviation were calculated for each
measure.
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
B) Chemistry Methods
General information and procedures. Nuclear magnetic resonance spectra (proton and
carbon) were recorded on Bruker AC200 F, and 300 or 500 Advance spectrometers at the
University of A Coruña, using CDCl3, D2O and CD3OD as the solvents and internal
standards. Multiplicities of 13C signals were obtained by DEPT. Medium-pressure
chromatographic separations were carried out on silica gel 60 (230–400 mesh). Optical
rotations were determined on a JASCO DIP-1000 polarimeter, with Na (589 nm) lamp
and filter. LREIMS and HRESIMS were measured on Applied Biosystems QSTAR Elite.
HPLC separations were carried out on an Agilent HP1100 liquid chromatography system
equipped with a solvent degasser, quaternary pump and an UV detector (Agilent
Technologies, Waldbronn, Germany). In the HPLC separations a Discovery® column HS
F5 (100x4.6 mm, 5 µm) and an Eclipse® column XDB-C18 (150x4.6 mm, 5 µm) were
used.
All moisture-sensitive reactions were carried out under an atmosphere of argon in flame-
dried glassware closed by rubber septum, unless otherwise noted. Solvents were distilled
prior to use under argon atmosphere and dried according to standard procedures using the
following desiccants: Na/benzophenone for THF and Et2O; CaH2 for dichloromethane,
pyridine and triethylamine; magnesium for methanol and anhydrous CaSO4 for acetone.
DMF was distilled over CaH2 and was kept over molecular Sieves 4Å under argon
atmosphere. Solutions and solvents were added via syringe or cannula. Thin layer
chromatography was performed using silica gel GF-254 Merck, spots were revealed
employing UV light (254 nm) and/or by heating the plate pre-treated with an ethanolic
27
ACS Paragon Plus Environment

Page 29 of 46
ACS Chemical Biology
1
2
3
4
5
6
7
8
solution of phosphomolibdic acid, a solution of cerium sulphate or a solution of
ninhydrine in BuOH-AcOH-H2O. CRYOCOOL apparatus was used for low-temperature
reactions. Abbreviations for the reagents used in the synthesis: TBTU: O-(Benzotriazol-
9
1-yl)-N,N,N′, N′-tetramethyluronium tetrafluoroborate; EDC: N-(3-Dimethylamino-
propyl)-N′-ethylcarbodiimide hydrochloride; HOBt 1-Hydroxybenzotriazole; CDI: 1,1
′- Carbonyldiimidazole.
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
Experimental Procedures and Physical Data of Compounds
Synthesis of acinetobatin analogue ent-Acb (1b)
Compound 14.- To a stirrer solution of acid 12 (1.00 g, 4.21 mmol) in DMF (20 mL)
was added amine 13 (0.80 g, 3.82 mmol), TBTU (0.80 g, 3.82 mmol) and Et3N (1.33 mL,
9.55 mmol), at room temperature overnight. The mixture was diluted with EtOAc (50
mL) and washed with NaHCO3 saturated aqueous solution, water, citric acid (0.1 M),
water and brine. The organic layer was dried over anhydrous MgSO4 and filtered, the
solvent was removed under reduced pressure. The crude was purified by silica gel
chromatography (hexane/EtOAc, 4:1), affording compound 14 as an amorphous solid
(1.49 g, 64%). 1H NMR (300 MHz, CDCl3): δH = 8.94 (d, J = 8.3 Hz, 1H, NH); 7.73 (dd,
J = 7.5, 2.1 Hz, 1H, H-6); 7.44-7.29 (m, 5H); 7.14-6.97 (m, 2H, H-4/H-5); 5.23 (s, 2H, –
OCH2Ph); 4.95-4.74 (m, 1H,–OCHMe2); 4.95-4.74 (m, 1H, H-8); 4.57 (hept, J = 6.1 Hz,
1H, –OCHMe2); 4.48-4.31 (m, 1H, H-9); 1.42-1.31 (m, 9H, –OCHMe2 & Me); 1.32-1.19
(m, 6H, –OCHMe2). 13C NMR (75 MHz, CDCl3): δC = 170.9 (CO, C-7); 166.3 (CO, C-
10); 150.8 (C, C-2); 146.7 (C, C-3); 135.3 (C); 128.6 (CH); 128.4 (CH); 128.3 (CH);
127.3 (C, C-1); 123.4 (CH, C-4); 123.3 (CH, C-5); 119.3 (CH, C-6); 71.3 (CH, –
OCHMe2); 68.5 (CH, –OCHMe2); 67.1 (CH, C-9); 58.1 (CH, C-8); 22.1 (CH3, Me); 22.0
(CH3, –OCHMe2); 21.5 (CH3, –OCHMe2); 20.1 (CH3, –OCHMe2). HRMS (ESI+): calcd
for C24H31NO6Na [M+Na]+ 452.2043, found 452.2041. [α]D25 = +27.3 (c = 0.52, CHCl3).
Compound 15.- SOCl2 (0.74 mL, 10.12 mmol) was added dropwise over a previously
prepared solution of compound 14 (0.72 g, 1.68 mmol) in CH2Cl2 (20 mL), cooled at 0
ºC. The reaction was allowed to reach room temperature and then, it was stirred overnight.
The reaction flask was placed on a water with ice bath and it was neutralized pouring a
saturated solution of NaHCO3 dropwise. After extraction with CH2Cl2 (3 x 50 mL), the
combined organic layers were dried over anhydrous MgSO4, filtered and the solvent
28
ACS Paragon Plus Environment