Mimicking salmochelin S1 and the interactions of its Fe(III) complex with periplasmic iron siderophore binding proteins CeuE and VctP

Article abstract of DOI:10.1016/j.jinorgbio.2018.10.008

A mimic of the tetradentate stealth siderophore salmochelin S1, was synthesised, characterised and shown to form Fe(III) complexes with ligand-to-metal ratios of 1:1 and 3:2. Circular dichroism spectroscopy confirmed that the periplasmic binding proteins

Full text of DOI:10.1016/j.jinorgbio.2018.10.008

JournalofInorganicBiochemistry190(2019)75–84
Contents lists available at ScienceDirect
Journal of Inorganic Biochemistry
journal homepage: www.elsevier.com/locate/jinorgbio
Mimicking salmochelin S1 and the interactions of its Fe(III) complex with
periplasmic iron siderophore binding proteins CeuE and VctP☆
Ellis J. Wildea,b, Elena V. Blagovaa, Thomas J. Sandersonb, Daniel J. Rainesb, Ross P. Thomasb,
⁎
⁎
Anne Routledgeb, Anne-Kathrin Duhme-Klairb, , Keith S. Wilsona,
a
Structural Biology Laboratory, Department of Chemistry, University of York, Heslington, York YO10 5DD, UK
b
Department of Chemistry, University of York, Heslington, York YO10 5DD, UK
A R T I C L E I N F O
A B S T R A C T
Keywords:
A mimic of the tetradentate stealth siderophore salmochelin S1, was synthesised, characterised and shown to
form Fe(III) complexes with ligand-to-metal ratios of 1:1 and 3:2. Circular dichroism spectroscopy confirmed
that the periplasmic binding proteins CeuE and VctP of Campylobacter jejuni and Vibrio cholerae, respectively,
bind the Fe(III) complex of the salmochelin mimic by preferentially selecting Λ-configured Fe(III) complexes.
Intrinsic fluorescence quenching studies revealed that VctP binds Fe(III) complexes of the mimic and structu-
rally-related catecholate ligands, such as enterobactin, bis(2, 3-dihydroxybenzoyl-L-serine) and bis(2, 3-dihy-
droxybenzoyl)-1, 5-pentanediamine with higher affinity than does CeuE. Both CeuE and VctP display a clear
preference for the tetradentate bis(catecholates) over the tris(catecholate) siderophore enterobactin. These
findings are consistent with reports that V. cholerae and C. jejuni utilise the enterobactin hydrolysis product bis(2,
3-dihydroxybenzoyl)-O-seryl serine for the acquisition of Fe(III) and suggest that the role of salmochelin S1 in
the iron uptake of enteric pathogens merits further investigation.
Salmochelin
Stealth siderophore mimic
Iron uptake
Vibrio cholerae
Periplasmic binding protein
1. Introduction
enterobactin derivatives, and siderophores with structural differences
cannot be bound [9,16,17]. Such stealth siderophores are therefore able
Iron is an essential element required for the survival and growth of
bacteria within a host organism. Iron(II) is in scarce supply in aerobic
environments, so iron(III) must be taken up, despite its poor aqueous
solubility of 10−18–10−17 mol dm−3 at pH 7 [1]. Requiring con-
centrations between 10−8 and 10−6 mol dm−3 for growth [2], complex
strategies must be employed by bacteria for acquisition of sufficient
iron for survival [1,3]. Siderophores are low molecular weight
(< 1500 Da) compounds with high affinity for iron(III) that are pro-
duced, secreted and employed by bacteria for iron(III) uptake [2,4].
Vibrio cholerae acquires essential iron by producing the catecholate
siderophore vibriobactin, and by poaching enterobactin and its linear
hydrolysis products from competing bacteria in the surrounding en-
vironment (Fig. 1) [5–8].
to evade the host immunoprotein siderocalin, allowing for more effec-
tive iron acquisition [16–19]. Examples include the salmochelin side-
rophores, which are C5-glucosylated analogues of enterobactin, and its
hydrolysis products, produced by pathogenic strains of Escherichia coli
(Fig. 2) [20–23]. They have high iron(III) affinity, but are more hy-
drophilic than enterobactin, with the glucose units providing steric bulk
[24,25]. These two properties allow the siderophores to avoid seques-
tration by siderocalin [26,27].
V. cholerae can colonise the mammalian gut, where enterobactin
and vibriobactin are sequestered by siderocalins [15–18,28]. Due to the
high pathogenicity of Vibrio cholerae, estimated to cause at least
120,000 global deaths per year [29,30], we were particularly interested
to investigate whether this species is capable of employing salmochelin-
type stealth siderophores.
In response, some host organisms produce siderocalins which can
bind a number of iron(III)-bound siderophores, including enterobactin
and vibriobactin, with subnanomolar dissociation constants, and hinder
bacterial iron(III)-siderophore uptake [9–15]. However, due to the
specificity of the siderocalin binding pocket, some functionalised
While a number of previous studies on the use of iron(III)-en-
terobactin in V. cholerae did not consider the potential presence of
enterobactin hydrolysis products, such as tris(2, 3-dihydroxybenzoyl-L-
serine) (trisDHBS), bis(2, 3-dihydroxybenzoyl-L-serine) (bisDHBS) and
☆
Dedicated to Prof. Bernt Krebs on the occasion of his 80th birthday.
⁎
Corresponding authors.
E-mail addresses: anne.duhme-klair@york.ac.uk (A.-K. Duhme-Klair), keith.wilson@york.ac.uk (K.S. Wilson).
URL: https://www.york.ac.uk/chemistry/ (K.S. Wilson).
https://doi.org/10.1016/j.jinorgbio.2018.10.008
Received 23 July 2018; Received in revised form 15 October 2018; Accepted 16 October 2018
Availableonline19October2018
0162-0134/©2018ElsevierInc.Allrightsreserved.

E.J. Wilde et al.
JournalofInorganicBiochemistry190(2019)75–84
OH
OH
OH
OH
O
O
OH
OH
HO
O
O
H
N
HN
HN
OH
O
OH
OH
O
OH
O
OH
O
O
O
HN
O
OH
OH
O
O
NH
O
NH
O
HO
HO
HO
HO
2, 3-
dihydroxy
benzoyl-
L-serine,
DHBS
Tris(2, 3-
dihydroxybenzoyl-L-
serine),TrisDHBS
Bis(2, 3-
dihydroxybenz
oyl-L-serine),
BisDHBS
Fig. 1. Catecholate siderophores used by Vibrio cholerae for iron(III)-uptake [7].
2, 3-dihydroxybenzoyl-L-serine (DHBS), recent studies give a more
complete picture of enterobactin-derived iron(III) acquisition
[7,31–34]. Whilst it remains unclear whether or not intact cyclic en-
terobactin can support growth of V. cholerae, it is now accepted that
iron(III)-trisDHBS and iron(III)-bisDHBS provide efficient iron(III) de-
livery pathways [34,35]. TonB, a protein that enables the active
transport of essential nutrients, is required for the transport of iron(III)-
siderophore complexes across the outer membrane via the two outer
membrane receptor proteins VctA and IrgA for linear enterobactin de-
rivatives [36], and the receptor for vibriobactin ViuA (Fig. 3)
[31,37,38]. Both periplasmic-binding protein dependent ABC trans-
porters VctPDGC and ViuPDGC have been shown to transport iron(III)-
vibriobactin and iron(III)-enterobactin derivatives into the cytoplasm.
Crystal structures have been reported for periplasmic binding proteins
VctP and ViuP: VctP in the apo form (PDB ID: 3TEF), and ViuP with
vibriobactin bound (PDB ID:3R5T) [39,40].
to establish the siderophore-binding properties of VctP, to further un-
derstand its exact role in iron-uptake in V. cholerae. A number of ca-
techolate siderophores and siderophore mimic compounds were
screened for their binding affinity with VctP. A tetradentate sal-
mochelin mimic was synthesised, its iron-binding properties char-
acterised, and its interaction with CeuE and VctP evaluated. This study
aimed to reveal whether these periplasmic binding proteins could be
involved in uptake of Fe(III)-salmochelin S1 (Fig. 2). This is the first
indication that V. cholerae may employ salmochelin-type siderophores
as an iron-uptake strategy that evades host siderocalin.
2. Materials and methods
2.1. Siderophores and siderophore mimics
Enterobactin was obtained from EMC microcollections (Tübingen)
and used as supplied after compound purity was confirmed by HPLC
and extinction coefficient determination. Bis(2, 3-dihydroxybenzoyl-L-
serine) (bisDHBS) and bis(2, 3-dihydroxybenzoyl)-1, 5-pentanediamine
(5-LICAM) were synthesised as previously reported [42,43].
Our previous studies on the periplasmic binding protein CeuE from
Campylobacter jejuni, which binds the tetradentate siderophore iron(III)-
bisDHBS, revealed a strong similarity between CeuE and VctP [41,42].
VctP shares sequence similarities (25.3% identical, 47.6% similar re-
sidues) and high structural similarities (r.m.s.d of 1.78 Å over 263 Cα
positions, 100% match for secondary structure elements PDB ID: 3ZKW,
3TEF) with CeuE. Both proteins contain conserved histidine and tyr-
osine residues that complete coordination of the iron(III) centre of
tetradentate siderophore complexes in CeuE [41–43]. It is proposed
that VctP has an analogous function to CeuE, and is optimised for
binding tetradentate iron(III)-siderophore complexes. We here proceed
2.2. Instrumentation
NMR spectra were recorded on Jeol EX and ES 400 MHz instru-
ments. 1H experiments were run at 399.78 MHz. 13C experiments were
run at 100.53 MHz and were proton decoupled. Positive and negative
ion electrospray ionisation mass spectrometry (ESI-MS) was performed
76

E.J. Wilde et al.
JournalofInorganicBiochemistry190(2019)75–84
Fig. 2. Salmochelin siderophores [24].
on a Bruker microTOF electrospray mass spectrometer. Elemental
analysis was performed on an Exeter CE-440 elemental analyser.
Infrared (ATIR) spectra were recorded on a Perkin Elmer FT-IR spec-
trum two spectrophotometer at ambient temperature. Melting points
were determined using a Stuart Scientific SMP3 melting point appa-
ratus. Specific rotation was recorded on a Jasco DIP-370 digital po-
larimeter. Fluorescence spectra were recorded on a Hitachi F-4500
fluorescence spectrophotometer at ambient temperature. Electronic
absorption spectra were recorded on a Shimadzu UV-1800 spectro-
photometer at ambient temperature. Circular dichroism was performed
on a Jasco J810 CD spectropolarimeter at 20 °C under a constant flow of
nitrogen.
7.72 mmol, 73%) Rf = 0.50 (1:4 ethyl acetate: petroleum ether
40–60 °C) M.P = 109–111 °C. 1H NMR: (400 MHz, CDCl3) δ: 10.99 (s,
H9, 1H); 7.76 (d, J = 2.0 Hz, H4/6, 1H); 7.25 (d, J = 2.0 Hz, H4/6,
1H); 3.96 (s, H10, 3H); 3.89 (s, H8, 3H)·13C NMR: (100 MHz, CDCl3) δ:
169.6 (C7); 152.0, 149.3, 114.2 (C1, 2, 3); 129.5, 124.9 (C4,6); 56.4
(C8); 52.7 (C10). HRMS (ESI): Calcd. [M + H]+ (C9H9IO4) m/
z = 308.9618; Obs. [M + H]+ m/z = 308.9629, Mean err 3.7 ppm.
Calcd. [M + Na]+ (C9H9INaO4) m/z = 330.9438; Obs. [M + Na]+ m/
z = 330.9446, Mean err 2.9 ppm. IR ATIR (cm−1): 3090 br w (CeH),
2943 w (CeH), 1674 m (C]O), 1607 m (C]C ar), 1471 s (CeH),
1343 m (OeH).
2.3.2. Methyl-5-iodo-3-hydroxysalicylate (3)
2.3. Synthesis of salmochelin mimics
Methyl-5-iodo-3-hydroxysalicylate was prepared by adaptation of a
literature protocol [49]. Methyl-5-iodo-3-methoxysalicylate (2.000 g,
6.49 mmol) was dissolved in anhydrous dichloromethane (10 mL) and
the mixture was stirred whilst the reaction flask was purged with N2.
BBr3 (1 M in dichloromethane, 12 mL, 12.00 mmol) was added drop-
wise with vigorous stirring. The reaction flask was then purged again
with N2 and the mixture stirred overnight at room temperature. The
reaction mixture was opened to air, cold H2O (20 mL) was carefully
added and the reaction mixture stirred for 1 h. The resulting pale pink
turbid mixture was dissolved in methanol (20 mL) and the solvent re-
moved in vacuo yielding a red residue. Methanol (10 mL × 3) was
added to the residue, the residue dissolved, and the solvent removed in
vacuo. The resulting peach coloured solid was redissolved in methanol
Methoxy-5-aceto-β-ᴅ-glucosyl-3, 4-benzyloxysalicylate (5) [44,45],
2, 3-bis(benzyloxy)benzoic acid (9) [46] and 2, 3-bis(benzyloxy)ben-
zoic acid N-hydroxysuccinimide ester (10) [47] were prepared as
documented in the literature. Compound structures with the atom
numbering scheme used for the NMR assignments are available in the
Supporting Information.
2.3.1. Methyl-5-iodo-3-methoxysalicylate (2)
Methyl-5-iodo-3-methoxysalicylate was prepared as in the litera-
ture, replacing the o-vanillin substrate for 5 [48]. The crude product
was recrystallised from a minimum amount of hot ethanol. (2.379 g,
77

E.J. Wilde et al.
JournalofInorganicBiochemistry190(2019)75–84
Fig. 3. Known functions of proteins involved in Fe(III) ca-
techolate siderophore uptake in V. cholerae. Uptake through
the outer membrane is mediated by the outer membrane re-
ceptors VctA, IrgA and ViuA, which are energised by the
TonB-ExbB-ExbD complex. Periplasmic binding proteins
(VctP, ViuP) capture the Fe(III) siderophore complexes in the
periplasm for subsequent uptake through the inner membrane
transporters VctDG, VctC, ViuDG and ViuC. In the cytoplasm,
ViuB facilitates the release of the iron form its siderophore
complex via reduction of Fe(III) to Fe(II). Ent = enterobactin,
Vib = vibriobactin, trisDHBS = tris(2, 3-dihydroxybenzoyl-L-
serine), bisDHBS = bis(2, 3-dihydroxybenzoyl-L-serine).
(50 mL) and concentrated H2SO4 (2 mL) and heated under reflux
overnight. The reaction mixture was cooled to room temperature and
the solvent removed in vacuo, yielding a colourless oil and a white solid.
The residue was dissolved in ethyl acetate (150 mL) and was washed
with saturated NaHCO3 (3 × 90 mL) and brine (2 × 90 mL). The or-
ganic portion was dried over MgSO4, filtered and the solvent removed
in vacuo yielding a white solid (1.710 g, 5.82 mmol, 89%) Rf = 0.47
(1:4 ethyl acetate: petroleum ether 40–60 °C) M.P = 136–138 °C (Lit.
133–134 °C). HRMS (ESI): Calcd. [M + Na]+ (C9H9INaO4) m/
z = 316.9281; Obs. [M + Na]+ m/z = 316.9282, Mean err 0.5 ppm.
Calcd. [M + 2Na]+ (C9H9INa2O4) m/z = 338.9101; Obs. [M + 2Na]+
m/z = 338.9099, Mean err 0.2 ppm. 1H NMR, 13C NMR, HRMS (ESI):
consistent with literature data [44].
2.3.4. Methoxy-5-benzyloxy-β-ᴅ-glucosyl-3, 4-benzyloxysalicylate (6)
Methoxy-5-benzyloxy-β-ᴅ-glucosyl-3, 4-benzyloxysalicylate was
prepared based on a literature protocol [44,45]. Methoxy-5-aceto-β-ᴅ-
glucosyl-3, 4-benzyloxysalicylate (1.314 g, 1.94 mmol) was dissolved
in dry methanol (100 mL) and Na2CO3 (1.026 g, 9.68 mmol) was
added. The resulting suspension was stirred under reflux at 65 °C
overnight, then reduced in vacuo to a pale brown solid. The solid was
transferred to a Schlenk tube, and NaH (533 mg, 60% in mineral oil,
13.3 mmol) was added. The solids were dried in vacuo for 2 h, before
the addition of DMF (25 mL) and cooling of the solution to 0 °C. Bu4NI
was dried in vacuo for 15 min, to which, benzyl bromide (3.00 mL,
4.320 g, 25.26 mmol) and DMF (5 mL) were added. The solution was
cooled to 0 °C for 5 min, and was then added dropwise to the methyl-5-
aceto-β-ᴅ-glucosyl-3,4-benzyloxysalicylate solution over 5 min at 0 °C.
After 5 min of stirring at 0 °C, the pale brown reaction mixture was
stirred at RT under N2 overnight. The reaction mixture was opened to
air, and deionised water (40 mL) carefully added. The resulting solu-
tion was extracted with ethyl acetate (3 × 50 mL). The organic por-
tions were combined and dried over MgSO4, filtered and solvent re-
moved in vacuo to yield a white solid. The solid was purified by
column chromatography 20% ethyl acetate in petroleum ether
40–60 °C yielding the product as a white solid (0.719 g, 0.82 mmol,
42%) Rf = 0.20 (20% ethyl acetate in petroleum ether 40–60 °C)
M.P = 84–86 °C. HRMS (ESI): Calcd. [M + Na]+ (C56H54NaO9) m/
z = 893.3660; Obs. [M + Na]+ m/z = 893.3655, Mean err 0.6 ppm.
1H NMR, 13C NMR, HRMS (ESI): consistent with literature data
[44].
2.3.3. Methyl-5-iodo-3, 4-benzyloxysalicylate (4)
Methoxy-5-benzyloxy-5-iodo-3, 4-benzyloxysalicylate was prepared
based on a literature protocol [44,45]. The residue was dissolved in 1:1
chloroform: ethyl acetate (10 mL) and purified by flash column chro-
matography (1:9 ethyl acetate: petroleum ether 40–60 °C) followed by
recrystallisation in a minimum amount of chloroform and an excess of
petroleum ether 40–60 °C. The white needle crystals were collected and
dried in vacuo (1.613 g, 3.40 mmol, 78%) Rf = 0.37 (1:9 ethyl acetate:
petroleum ether 40–60 °C) M.P = 103–105 °C (Lit. 105–106 °C). HRMS
(ESI): [M + K]+ (C9H9INKO4) m/z = 512.9960; Obs. [M + K]+ m/
z = 512.9957, Mean err 0.1 ppm. 1H NMR, 13C NMR, HRMS (ESI):
consistent with literature data [44].
78

E.J. Wilde et al.
JournalofInorganicBiochemistry190(2019)75–84
2.3.5. Benzyloxy-5-benzyloxy-β-ᴅ-glucosyl-3, 4-benzyloxysalicylate (7)
Benzyloxy-5-benzyloxy-β-ᴅ-glucosyl-3, 4-benzyloxysalicylate was
prepared by the same method as methyl-5-benzyloxy-β-ᴅ-glucosyl-3,4-
benzyloxysalicylate. The compounds were then separated by column
chromatography yielding the product as a white solid. (0.381 g,
0.40 mmol, 21%) Rf = 0.26 (1:4 ethyl acetate: petroleum ether
40–60 °C) M.P = 104–106 °C. 1H NMR: (400 MHz, CDCl3) δ: 7.52 (d,
H6, J = 1.7 Hz, 1H) 7.40–7.28 (m, H10, H15, H26–28, H31–33,
H36–38, H41–43, H46–48, 29H); 7.26–7.21 (m, H4, H11, H16, 5H)
6.94–6.92 (m, H12, H17, 2H) 5.34 (d, H44, J = 12.4 Hz, 2H); 5.07 (d,
H8, J = 10.1 Hz, 2H); 4.98 (d, H13, J = 2.2 Hz, 2H); 4.97–4.87 (m,
H24/29, H39, 3H) 4.66–4.57 (m, H24/29/H23, 3H); 4.43 (d, H23,
J = 10.1 Hz 1H); 4.21 (d, H18, J = 9.5 Hz, 1H); 3.82–3.76 (m, H20/21/
24/29, H34, 5H); 3.61–3.59 (m, H22, 1H); 3.44 (t, H19, J = 9.0 Hz,
1H). 13C NMR: (100 MHz, CDCl3) δ: 165.9 (C7); 152.5 (C3); 148.0 (C2);
138.6 (C25); 138.2, 138.1, 137.4, 137.2, 136.3, 135.9, (C1/10/15/30/
35/41/45); 128.6, 128.5, 128.4, 128.4, 128.4, 128.3, 128.2, 128.2,
128.1, 128.1, 128.0, 127.9, 127.9, 127.9, 127.8, 127.8, 127.7, 127.7,
127.6, 127.6, 127.6, 126.4 (C5/10–12/15–17/26–28/31–33/36–38/
42–44/46–48); 122.0 (C6); 116.5 (C4); 86.6 (C20/21); 83.9 (C19); 80.9
(C18); 79.3 (C22); 78.2 (C20/21); 75.6 (C8); 75.1 (C24/29/34); 73.2
(C23); 70.9 (C39); 68.9 (C24/29/34); 68.1 (C13); 66.9 (C44). HRMS
(ESI): Calcd. [M + Na]+ (C62H58NaO9) m/z = 969.3973; Obs.
[M + Na]+ m/z = 969.3982, Mean err 1.3 ppm.
hexafluorophosphate (HATU) (0.268 g, 0.70 mmol) was added. The
solution was stirred for 1 h before the addition of N,N-diisopropy-
lethylamine (DIPEA) (184 μL, 0.137 g, 1.06 mmol) and 1-amino, 5-
(2, 3-dibenzyloxybenzamide) pentane (0.160 g, 0.35 mmol) and the
mixture stirred overnight. The resulting brown solution was reduced
in vacuo to a brown residue. The residue was purified twice by
column chromatography (1:2 ethyl acetate: chloroform) (1:4 ethyl
acetate: chloroform) yielding
a white solid product (0.204 g,
0.16 mmol, 55%) Rf = 0.36 (1:4 ethyl acetate: chloroform)
M.P = 96–98 °C. 1H NMR: (400 MHz, CDCl3) δ: 7.93 (d, H6,
J = 1.8 Hz, 1H); 7.90 (m, H18/24,1H); 7.87 (t, H18/24, J = 5.5 Hz,
1H); 7.76 (dd, H29/31, J = 6.4 Hz, J = 3.2 Hz, 1H); 7.50–7.14 (m,
H4/30/9–11/14–16/34–36/39–41/50–52/55–57/60–62/65–67,
42H); 6.98 (m, H29/31, 2H); 5.17 (s, H12, 2H); 5.11–5.01 (m, H32/
37, 2H); 5.08 (s, H32/37, 2H); 4.98 (br s, H7, 2H); 4.95–4.86 (m,
H48/53, 3H); 4.67–4.56 (m, H53/64, 3H); 4.45 (d, H47, J = 10.5 Hz,
1H); 4.26 (d, H42, J = 9.6 Hz, 1H); 3.86–3.73 (m, H47/44/58/46,
5H); 3.63 (dt, H45, J = 9.2 Hz, J = 3.2 Hz, 1H); 3.52 (t, H43,
J = 9.2 Hz, 1H); 3.25–3.18 (m, H19, H23, 4H); 1.30–1.23 (m, H20,
H22, 4H); 1.16–1.10 (m, H21, 2H). 13C NMR: (100 MHz, CDCl3) δ:
164.9 (C17/25); 164.7 (C17/25); 151.7 (C3/28); 151.4 (C3/28);
146.7 (C2/27); 146.3 (C2/27); 138.7, 138.3, 137.7, 136.4, 136.3,
136.3, 136.3, 135.5 (C8/13/33/38/49/54/59/64); 128.8, 128.7,
128.7, 128.6, 128.4, 128.4, 128.2, 128.2, 128.2, 128.0, 127.8, 127.7,
127.6, 127.6, 127.5, 127.5, 127.4, 127.2 (C9–11/14–16/34–36/
39–41/50–52/55–57/60–62/65–67); 124.4 (C4/29/30/31); 123.3
(C4/29/30/31); 122.2 (C4/29/30/31); 116.8(C29/31); 115.8 (C6);
86.7 (C44/46); 84.0 (C43); 81.1 (C42); 79.3 (C45); 78.3 (C44/46);
76.4 (C32/37); 75.6 (C32/37); 75.1 (C48/53); 74.9 (C47); 73.4
(C53/63); 71.2 (C32/37); 71.0 (C48/53); 69.1 (C58); 39.6 (C19/23);
39.5 (C19/23); 28.9 (C20/22), 28.9 (C20/22), 24.5 (C21). HRMS
(ESI): Calcd. [M + H]+ (C81H80N2Na1O11) m/z = 1279.5654; Obs.
[M + Na]+ m/z = 1279.5606, Mean err 1.9 ppm. IR ATIR (cm−1):
3384 w br (NeH), 3287 w br (NeH), 3063 w (CeH), 3030 w (CeH),
2920, 2859 w br (CeH), 1638 m (C]O), 1605 w (C]O), 1577 m
(C]C ar).
2.3.6. 1-Amino-5-(2,3-dibenzyloxybenzamide) pentane (11)
1, 5-Diaminopentane (1.427 g, 14.00 mmol) and triethylamine
(1.396 g, 13.80 mmol) were dissolved in THF (120 mL). A solution of 2,
3-bis(benzyloxy)benzoic acid N-hydroxysuccinimide ester (3.008 g,
6.97 mmol) in THF (60 mL) was added dropwise over 2 h and the
mixture was left to stir overnight. The solvent was removed in vacuo
yielding an off-white solid which was taken up in chloroform (120 mL)
and washed with NaHCO3 (100 mL), brine (100 mL) and 2.25:1 1 M
HCl: brine (130 mL). The organic layer was dried over MgSO4, filtered
and the solvent removed in vacuo. The residue was purified via silica
column chromatography (90:10:1 CHCl3: MeOH: NH3(aq)). (1.859 g,
59%) Rf = 0.18 (90:10:1 CHCl3: MeOH: NH3(aq)) M.P = 133–134 °C.
1H NMR: (400 MHz, CDCl3) δ: 7.99 (t, J = 5.5 Hz, H8, 1H); 7.73–7.71
(m, H6/4, 1H); 7.49–7.35 (m, H11–13, H16–18, 10H); 7.16 (d,
J = 2.3 Hz, H6/4, 2H); 7.14 (t, J = 4.6 Hz, H5, 1H); 5.16 (s, H9/14,
2H); 5.08 (s, H9/14, 2H); 3.27 (dd/q, J = 6.0 Hz J = 13.3 Hz, H19, 2H);
2.77 (t, J = 6.6 Hz, H23, 2H); 1.48 (dt, J = 7.3 Hz, H20, 2H); 1.37–1.20
(m, H21, H22, 4H). 13C NMR: (100 MHz, CDCl3) δ: 165.0 (C7); 151.7,
146.7, 136.4, 128.7, 128.6, 128.6, 127.7 (C1–3/10–13/15–18); 124.4
(C4/6); 123.3 (C4/6); 116.7 (C5);76.3, 71.2 (C9,14); 42.0 (C23); 39.6
(C19); 33.3, 29.1, (C20/22); 24.2 (C21). HRMS (ESI): Calcd. [M + H]+
(C25H29N2O3) m/z = 405.2173; Obs. [M + H]+ m/z = 405.2168,
Mean err 3.5 ppm. IR ATIR (cm−1): 3365 m (NeH), 307b m br (NeH),
2803 m br (CeH), 1641 m (C]O), 1571 m (C]C ar).
2.3.9. 5-β-ᴅ-glucosyl-5-LICAM (14)
5-β-ᴅ-Glucosyl-5-LICAM was prepared based on a literature protocol
[44,45]. 5-Benzyloxy-β-ᴅ-glucosyl-3, 4-benzyloxy-4-LICAM (0.200 g,
0.1590 mmol) was dissolved in toluene (1 mL) and ethanol added
(30 mL) followed by 3 spatula tips of Pd(OH)2 20% on carbon. The
system was purged with nitrogen before purging with hydrogen for
30 min. The reaction mixture was stirred under balloon pressure of
hydrogen for 18 h, and purged with nitrogen before opening to air. The
catalyst was removed by filtration, and the solvent removed in vacuo to
yield a pale colourless oil. The solid off white product was obtained by
cooling the obtained oil in liquid nitrogen and removing all residual
solvent in vacuo (0.086 g, 0.16 mmol, 90%). M.P = 180–182 °C. 1H
NMR: (400 MHz, MeOD) δ: 7.31 (s, H4/6, 1H); 7.20 (d, H20/22,
J = 8.2 Hz 1H); 7.01 (s, H4/6, 1H); 6.91 (d, H20/22, J = 7.8 Hz, 1H);
6.70 (t, H21, J = 7.8 Hz, 1H); 4.02 (d, H26, J = 9.2 Hz, 1H); 3.87 (d,
2.3.7. Benzyloxy-β-ᴅ-glucosyl-3, 4-benzyloxysalicylate (12)
5-Benzyloxy-β-ᴅ-glucosyl-3,4-benzyloxysalicylate was prepared as
in the literature [44,45]. Purification involved the addition of petro-
leum ether 40–60 °C (3 × 10 mL) and removal in vacuo to yield an off
white solid residue. The residue was purified by column chromato-
graphy (1:4 methanol: chloroform) yielding a white solid product
(0.359 g, 0.42 mmol, 87%) Rf = 0.56 (1:9 methanol: chloroform)
M.P = 97–99 °C. HRMS (ESI): Calcd. [M + Na]+ (C55H52NaO9) m/
z = 879.3504; Obs. [M + Na]+ m/z = 879.3522, Mean err 2.4 ppm.
1H NMR, 13C NMR, HRMS (ESI): consistent with literature data
[44].
2
3
H34, J = 10.5 Hz, 1H); 3.72 (dd, H34, JH34a/H34b = 11.9 Hz, JH34/
H30 = 5.0 Hz, 1H); 3.46–3.35 (m, H11/15/27–30, 8H); 1.72–1.64 (m,
H12/14, 4H); 1.50–1.44 (m, H13, 2H). 13C NMR: (100 MHz, MeOD) δ:
171.6 (C9/17); 171.6 (C9/17); 150.4 (C3/19); 150.3 (C3/19); 147.5
(C2/18); 147.1 (C2/18); 131.3 (C1/23); 119.7 (C21/20/22); 119.4
(C21/20/22); 118.8 (C4/6); 118.1 (C20/22); 116.9 (C4/6); 116.2 (C5);
83.3 (C26); 82.1 (C27/28/29/30); 79.8 (C27/28/29/30); 76.5 (C27/
28/29/30); 71.8 (C27/28/29/30); 63.0 (C34); 40.5 (C11/15); 40.5
(C11/15); 30.3 (C12/14); 30.3 (C12/14), 25.6 (C13). HRMS (ESI):
Calcd. [M + H]+ (C25H32N2O11) m/z = 535.1933; Obs. [M + H]+ m/
z = 535.1912, Mean err 3.5 ppm. IR ATIR (cm−1): 3288 s br (OeH),
2930 m br (CeH), 1641 m (C]O), 1589 m (C]C ar). Elemental
Analysis: Calcd. for [C25H32N2O11.0.9EtOH.1.2H2O]: %C 53.68, %H
6.70, %N 4.66; Measured for [C25H32N2O11.0.9EtOH.1.2H2O]: %C
2.3.8. 5-Benzyloxy-β-ᴅ-glucosyl-bis(3, 4-benzyloxy)-5-LICAM (13)
5-Benzyloxy-β-ᴅ-glucosyl-3,
4-benzyloxysalicylate
(0.252 g,
0.29 mmol) was dissolved in DMF (10 mL) to which 1-[Bis(di-
methylamino)methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium 3-oxid
79

E.J. Wilde et al.
JournalofInorganicBiochemistry190(2019)75–84
53.76, %H 6.47, %N 4.52. Specific Rotation: [α]D (Methanol, conc.
2.7. Circular dichroism spectroscopy
0.311 g/100 mL) + 5.7
Spectra were recorded in accordance with a published procedure
[43], between 300 nm and 700 nm with data pitch 0.5 mm in con-
tinuous scanning mode with speed of 100 nm/min and response of 2 s.
The bandwidth was set to 2 nm, with a path length of 1 cm. Spectra
were recorded five times and averaged.
2.4. Jobs plot
The iron-binding properties of 5-β-ᴅ-glucosyl-n-LICAM (n = 4, 5)
were investigated via a Job plot method [42]. 5-β-ᴅ-Glucosyl-n-LICAM
(n = 4, 5) was combined with varying concentrations of iron(III) to
establish preferred metal to ligand binding ratios. The ratios spanned
100% ligand to 100% iron(III), with 5% intervals. Between 60:40 and
50:50 ligand to iron(III) ratios, 2% intervals were used. This protocol
resulted in 24 samples over the full range. Stock solutions of 10 mM Fe
(III)-nitrilotriacetate (NTA) in H2O and 10 mM 5-β-ᴅ-glucosyl-n-LICAM
(n = 4, 5) in DMSO were used in the necessary ratios totalling 200 μL to
make up a 2 mL solution of each ratio in 1800 μL 0.11 M Tris-HCl
pH 7.5, 150 mM NaCl, resulting in an overall iron(III)-ligand con-
centration of 400 nM. DMSO was kept at 5% v/v for all final solutions.
A UV/Vis spectrum from 400 nm to 700 nm was run for each solution
after 1 h of equilibration. The spectra were rerun after 7 days of equi-
libration. λmax values were monitored at 495 nm and 555 nm and
plotted against ligand to protein ratio, and the maximum absorbance
for each wavelength recorded at the relevant ligand:protein ratio.
Solutions of CeuE and VctP were diluted to 2.5 × 10−3 M in 0.11 M
Tris-HCl pH 7.5, 150 mM NaCl buffer. A Fe(III)-ligand stock solution
containing equimolar NTA was made at a concentration of 5 × 10−4 M,
by adding 10 μL of both 10 mM ligand in DMSO and 10 μL 10 mM Fe
(III)-NTA in H2O to 180 μL 0.11 M Tris-HCl pH 7.5, 150 mM NaCl
buffer.
Spectra were recorded of solutions containing 880 μL 0.11 M Tris-
HCl pH 7.5, 150 mM NaCl buffer, 100 μL Fe(III)-ligand NTA stock so-
lution and 20 μL 2.5 × 10−3 M CeuE, resulting in a 1:1 ratio of ligand to
protein at 5 × 10−5 M. The spectrum of buffer was subtracted from all
spectra including the spectra for free ligand and apo protein.
3. Results and discussion
3.1. Salmochelin mimic synthesis
A salmochelin mimic, Sal-5-LICAM (14), was designed based on a
previously studied tetradentate siderophore mimic, 5-LICAM [43], and
synthesised as shown in Scheme 1. The synthesis of aryl C-glycosides,
such as (12), via metal-catalysed cross coupling reactions has previously
been described [44,45,50,51].
2.5. Cloning, expression and purification
CeuE was prepared as previously described [41,43]. The VctP gene
for truncated VctP (residues 51–333), with signal peptide removed, was
amplified by PCR from the synthetic DNA purchased from Thermo-
Fisher scientific. Primers used were of the following sequence: FWD_51
TCCAGGGACCAGCAGAGACAGTAACGATTGAACATCGCTTG, REV_333
To obtain the iodinated intermediate (4) for the cross coupling step,
commercially available (1) was selectively iodinated in the C5 position
using iodine monochloride, silver nitrate and pyridine to yield (2) [48],
then demethylated via BBr3 deprotection (3), followed by benzyl ether
protection of both phenolic hydroxyl groups to generate (4) in 51% yield.
The acetate protected β-ᴅ-glucose unit was then installed at the C5 po-
sition via Negishi coupling of (4) with acetobromo-α-ᴅ-glucose, using a
nickel catalyst and conditions adapted from a literature procedure [44].
The activation of (4) using zinc, lithium and iodine in DMF reached
completion within an hour and if the reaction mixture was left for 12 h,
as recommended in the literature [44], decomposition occurred. By
performing the subsequent Negishi coupling in a glovebox to avoid ex-
posure to the atmosphere, a yield of 66% of (5) was obtained [44,45].
The acetate protecting groups on the glucose unit were replaced with
benzyl groups, yielding two products (6 and 7), which were combined
and exposed to sodium hydroxide to afford (12) in 87% yield. Further
details and reaction conditions are provided in the experimental section.
In addition, compound 11 was synthesised, to allow for the selective
generation of a mono-glucosylated salmochelin mimic. (8) was trans-
formed into (9) via standard oxidation and benzyloxy protection reac-
tions [46]. The carboxylic acid of (9) was activated via production of
the N-hydroxysuccinimide ester (10) [47], which was then coupled
with 1, 5-diaminopentane to produce amide bonded (11). The required
second amide bond was formed between compounds (11) and (12) via a
standard HATU coupling reaction yielding (13). Global deprotection of
benzyl groups via standard hydrogenolysis afforded the target sal-
mochelin-inspired siderophore mimic Sal-5-LICAM (14).
TGAGGAGAAGGCGCGTTACTGCATACCAACTGACGCTTTCATG.
The
gene product was directionally cloned into a Lic-adapted pET 28a
vector (YSBLic3C), containing a 3C protease cleavable N-terminal hexa-
histidine tag, by the In-Fusion method (Clontech). Protein was ex-
pressed in E. coli strain BL21(DE3). Cells were grown with shaking at
37° in Luria-Bertani broth media containing 30 μg mL−1 kanamycin and
induced with 1 mM isopropyl β-D-thiogalactopyranoside at an OD600 of
0.6–0.8. The cells were harvested by centrifugation 4 h after induction
and resuspended in buffer A (50 mM Tris-HCl, pH 7.8, 500 mM NaCl,
10 mM imidazole) in the presence of protease inhibitor cocktail and
lysed by sonication on ice. The soluble crude extract was collected by
centrifugation at 19,900 rpm. The standard 3-step procedure was used
for VctP purification. Initially a 5 mL His-Trap chelating column
(Amersham Pharmacia) was charged with nickel and equilibrated with
buffer A. Fractions containing VctP were eluted with buffer B (50 mM
Tris-HCl, pH 7.8, 500 mM NaCl, 500 mM imidazole). 3C protease for
removal of the histidine tag was added in a 1:50 ratio to an overnight
dialysis against buffer C (50 mM Tris-HCl, pH 7.5, 150 mM NaCl). The
cleaved sample was loaded in buffer C onto a second His-Trap chelating
column, and eluted with buffer B. Non-tagged protein was eluted in the
flow-through (FT) fractions which were combined, concentrated by
centrifugal ultrafiltration (Amicon Ultra) and purified by gel filtration
on a Superdex S200 column in buffer D (50 mM Tris-HCl, pH 8.0;
150 mM NaCl). The final pure sample was concentrated to 90 mg mL−1
.
The molecular mass of the protein was confirmed by electro-spray mass
spectrometry.
3.2. Iron binding by method of continuous variation
2.6. Fluorescence quenching titrations
Sal-5-LICAM (14) was complexed with iron(III) in varying ligand:
iron(III) ratios, and the absorbance spectra recorded (Fig. 4A). The λmax
was observed to shift from around 495 nm to around 555 nm as the Fe
(III) to ligand ratio was increased. Plotting the absorbance at 555 nm and
495 nm for each spectrum over the range of ratios, gave the Job plot
shown in Fig. 4B. The absorbance at 555 nm peaked at a 50:50 ratio,
whilst the absorbance at 495 nm peaked at a 60:40 ratio. This indicated
that a mixture of 1:1 species and a 3:2 species was present in solution, as
Intrinsic fluorescence quenching titrations were carried out as pre-
viously described. Spectra were recorded with an excitation of 280 nm,
emission range of 295–410 nm, 10 nm excitation slit width, 10 nm
emission slit width, 60 nm/min scanning speed, automatic response,
corrected spectra and 700 V detector voltage using a Hitachi F-4500
fluorescence spectrophotometer [42,43].
80

E.J. Wilde et al.
JournalofInorganicBiochemistry190(2019)75–84
Scheme 1. Synthesis of Sal-5-LICAM (14) a: AgNO3, ICl, Pyridine, CHCl3 b: 1) BBr3, DCM 2) H2SO4, MeOH, 75 °C c: BnBr, NaI, DMF d: 1-Bromo-α-D-glucose
tetraacetate, Zn, LiCl, Ni(COD)2, tBuTerpy, DMF e: 1) Na2CO3, MeOH 2) BnBr, NaH, Bu4NI, DMF f: NaOH, THF, MeOH g: 1) BnCl, K2CO3, EtOH 2) H3NSO3, NaClO3,
Acetone, H2O h: N-hydroxysuccinimide, DCC, Dioxane i:1,5-diaminopentane, Et3N, THF j: HATU, Et3N, DMF k: H2, Pd(OH)2/C 10%, EtOH/Toluene.
represented by the schematic diagram in Fig. 4D. After 7 days, the
spectra were repeated, to establish whether further equilibration had
been taken place. For both wavelengths, the maximum absorbance was
achieved at a 60:40 ratio of Sal-5-LICAM: Fe(III) (Fig. 4C), which in-
dicated that the 3:2 species is thermodynamically more stable and pre-
dominates over time. Salmochelin siderophores are known to have a
reduced membrane partition coefficient when compared to their en-
terobactin equivalents [25]. It may therefore be that the hydrophilic
nature of the glucose moieties and their disposition to hydrogen bond in
aqueous solution allows the ligands to assemble and form a 3:2 species
with solvent exposed sugars shielding the hydrophobic backbones
[52–54]. However, considering the restricted iron levels within the host
organism combined with the need for fast bacterial iron uptake, the ki-
netic 1:1 species is probably biologically more relevant.
interactions with the chiral binding pocket of CeuE and the direct co-
ordination of two amino acid side chains, Tyr288 and His227, to the Fe
(III) centre. Hence, the CD spectra confirmed that CeuE is able to bind
the Fe(III) complex of the salmochelin S1 mimic and that CeuE shows a
preference for Λ-configured coordination centres.
A subsequent fluorescence quenching titration of CeuE with Fe(III)-
Sal-5-LICAM (Fig. 5B), gave a dissociation constant of 511
76 nM,
around 50 times larger than that for the binding of Fe(III)-5-LICAM or
the natural siderophore Fe(III)-bisDHBS with CeuE [42,43]. This sug-
gests that although CeuE is able to bind Fe(III)-Sal-5-LICAM, it appears
better suited to the binding of Fe(III)-bisDHBS than salmochelin S1.
3.4. Binding of the salmochelin-inspired iron(III)-siderophore mimic to VctP
The CD spectra obtained upon addition of Fe(III)-Sal-5-LICAM to
VctP also showed an induced Λ-configuration (Fig. 6A). As VctP con-
tains conserved histidine and tyrosine residues, it is likely that the Λ-
configured binding mode between Fe(III)-bound tetradentate side-
rophores is similar in VctP and CeuE. VctP was titrated with Fe(III)-Sal-
3.3. Binding of the salmochelin mimic to CeuE
The interactions of the Fe(III) complex of the salmochelin mimic
with CeuE were studied by circular dichroism (CD) spectroscopy
(Fig. 5A). In the absence of protein, no significant signal was observed
in the wavelength range of the ligand-to-metal charge transfer band,
indicating that a racemic mixture of Λ- and Δ-configured Fe(III) centres
was present in solution. However, the CD spectrum obtained in the
presence of CeuE shows clear negative bands between 310 and 540 nm
and a weak positive band above 550 nm. These bands are indicative of
Fe(III)-catecholamide complexes that adopt the Λ-configuration. A si-
milar chiral preference was previously reported for similar bis(ca-
techolate) siderophores and mimics [41–43] and shown by X-ray
crystallography to originate from H-bonding and electrostatic
5-LICAM in triplicate giving a dissociation constant of 9.4
3.0 nM
(Fig. 6B). Thus VctP binds salmochelin S1 with high affinity, which
suggests that V. cholerae is able to use the salmochelin stealth side-
rophores for Fe(III)-uptake via the VctPDGC system.
3.5. Comparison of the iron(III)-siderophore binding properties of VctP with
those of CeuE
Dissociation constants were measured for VctP titrated with Fe(III)-
enterobactin, Fe(III)-bisDHBS, Fe(III)-5-LICAM and compared with
81

E.J. Wilde et al.
JournalofInorganicBiochemistry190(2019)75–84
Fig. 4. Selected electronic absorbance spectra obtained at the indicated metal-
to-ligand ratios; [M] + [L] = 0.4 mM; 0.1 M Tris-HCl; 150 mM NaCl; 5% v/v
DMSO; pH 7.5 (A), Job plots (B: 1 h C: 7 days) for the reaction of Sal-5-LICAM
with iron(III), and schematic representation of a 1:1 and 3:2 tetradentate li-
gand: iron complex (D) (solvent molecules are depicted as Sol).
Fig. 5. A: CD spectra for Fe(III)-Sal-5-LICAM, showing some induced Λ-con-
figuration upon complexation with CeuE. B: Fluorescence quenching titration
curves (in triplicate) obtained with CeuE and Fe(III)-Sal-5-LICAM.
those obtained for CeuE (Table 1). VctP bound Fe(III)-enterobactin with
a dissociation constant of 369
25 nM, and with Λ-configured com-
plexes dominating in solution. This affinity was around ten times
stronger than that estimated for the binding of CeuE to Fe(III)-en-
terobactin. The binding of VctP to both Fe(III)-bisDHBS and Fe(III)-5-
LICAM was too strong to be quantified via fluorescence quenching ti-
trations, due to reaching the limits of detection, and were both esti-
mated to be in the sub-nanomolar range. These results suggest that VctP
is able to bind tetradentate catecholate siderophores with very high
affinity, including the salmochelin S1 mimic Sal-5-LICAM, but may also
be able to use enterobactin for iron uptake in V. cholerae. The ability to
use both tetradentate and hexadentate catecholate siderophores, as well
as stealth siderophores, may contribute to the virulence of V. cholerae.
O
Fe
O
O
Fe
O
O
O
O
O
O
O
O
O
O
Fe
O
O
4. Summary and conclusions
O
Sol
Sol
The salmochelin S1 mimic Sal-5-LICAM was synthesised, char-
acterised and shown to form both 1:1 and 3:2 complexes if equilibrated
for 1 h ([L] + [M] = 0.4 mM). Similar iron(III)-binding stoichiometries
1 : 1
3 : 2
82

E.J. Wilde et al.
JournalofInorganicBiochemistry190(2019)75–84
LICAM and Sal-5-LICAM, over the tris(catecholate) enterobactin. In
case of CeuE, the structural basis for this preference for tetradentate
siderophores originates from two protein side chains (Tyr288 and
His227) that coordinate directly to the Fe(III) centre [41–43], whilst the
coordination of these side chains to Fe(III) enterobactin is not possible,
due to a lack of available coordination sites. The observed preference of
tetradentate siderophores is also consistent with the report that V.
cholerae utilises the enterobactin hydrolysis product bisDHBS for the
acquisition of Fe(III), but not intact enterobactin [34]. and the finding
that bisDHBS is a functional Fe(III) uptake mediator in C. jejuni [55].
Our results suggest that the role of the tetradentate stealth side-
rophore salmochelin S1 in the iron uptake of enteric pathogens merits
further investigation as it may be of importance in iron cross-feeding
within environments that are shared with salmochelin-producing E. coli
strains, for example inflamed mammalian small intestines [56]. Sal-
mochelin S1 utilisation may provide scavenging pathogenic species
with a competitive advantage in microbiota where the Fe(III) com-
plexes of enterobactin and vibriobactin are being removed by the innate
defence protein siderocalin, whilst glucosylated stealth siderophore,
such as salmochelin S1, evade capture [15–18,28].
Acknowledgements
Funding: We would like to thank the EPSRC (grant no. EP/
L024829/1, and studentship for E.J.W.) for support.
Competing interests
The authors declare no competing interests.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.jinorgbio.2018.10.008.
Fig. 6. A: CD spectra for Fe(III)-Sal-5-LICAM, showing Λ-configuration upon
binding to VctP. B: Triplicate binding curves for fluorescence quenching titra-
tion of VctP with Fe(III)-Sal-5-LICAM.
References
[1] C. Ratledge, L.G. Dover, Annu. Rev. Microbiol. 54 (2000) 881–941.
[2] M.L. Guerinot, Annu. Rev. Microbiol. 48 (1994) 743–772.
[3] C. Wandersman, P. Delepelaire, Annu. Rev. Microbiol. 58 (2004) 611–647.
[4] R.C. Hider, X. Kong, Nat. Prod. Rep. 27 (2010) 637–657.
[5] J.R. Butterton, S.B. Calderwood, J. Bacteriol. 176 (1994) 5631–5638.
[6] E.E. Wyckoff, A.-M. Valle, S.L. Smith, S.M. Payne, J. Bacteriol. 181 (1999)
7588–7596.
Table 1
Dissociation constants for CeuE and VctP with a selection of Fe(III)-side-
rophores and Fe(III)-siderophore mimics, as determined by fluorescence
quenching titrations.
[7] E.E. Wyckoff, A.R. Mey, S.M. Payne, Biometals 20 (2007) 405–416.
[8] B.E. Allred, T. Hoette, K.N. Raymond, Abstr. Pap. Am. Chem. Soc. 239 (2010) 1.
[9] T.M. Hoette, R.J. Abergel, J.D. Xu, R.K. Strong, K.N. Raymond, J. Am. Chem. Soc.
130 (2008) 17584–17592.
Fe(III)-Ligand
CeuE Kd/nM
VctP Kd/nM
Enterobactin
bisDHBS
3500
10.1
300
369
< 1
< 1
9.4
25
[10] R.J. Abergel, M.C. Clifton, J.C. Pizarro, J.A. Warner, D.K. Shuh, R.K. Strong,
K.N. Raymond, J. Am. Chem. Soc. 130 (2008) 11524–11534.
[11] M.A. Holmes, W. Paulsene, X. Jide, C. Ratledge, R.K. Strong, Structure 13 (2005)
29–41.
3.8 (lit [42])
5-LICAM
< 10 (lit [43])
Sal-5-LICAM
511
76
3.0
[12] B.E. Allred, P.B. Rupert, S.S. Gauny, D.D. An, C.Y. Ralston, M. Sturzbecher-Hoehne,
R.K. Strong, R.J. Abergel, Proc. Natl. Acad. Sci. 112 (2015) 10342–10347.
[13] B.E. Allred, C. Correnti, M.C. Clifton, R.K. Strong, K.N. Raymond, ACS Chem. Biol. 8
(2013) 1882–1887.
40 mM Tris-HCl buffer, pH 7.5, 150 mM NaCl, titrations were performed as
independent triplicates, with the uncertainty in the average shown.
[14] G. Bao, M. Clifton, T.M. Hoette, K. Mori, S.-X. Deng, A. Qiu, M. Viltard, D. Williams,
N. Paragas, T. Leete, R. Kulkarni, X. Li, B. Lee, A. Kalandadze, A.J. Ratner,
J.C. Pizarro, K.M. Schmidt-Ott, D.W. Landry, K.N. Raymond, R.K. Strong,
J. Barasch, Nat. Chem. Biol. 6 (2010) 602–609.
of 1:1 and 3:2 were observed previously with other bis(catecholates),
such as 4-LICAM and bisDHBS [41,42]. Only after a further equilibra-
tion period of seven days was the thermodynamically more stable 3:2
complex seen to predominate. Initial protein-binding studies with dia-
stereoisomeric mixtures of the Fe(III) complexes were carried out by CD
spectroscopy and showed that both periplasmic binding proteins, CeuE
of C. jejuni and VctP of V. cholerae, preferentially select for Λ-configured
Fe(III) complexes of the salmochelin S1 mimic, as seen previously upon
binding of bis(catecholates) to CeuE [41–43].
[15] D.H. Goetz, M.A. Holmes, N. Borregaard, M.E. Bluhm, K.N. Raymond, R.K. Strong,
Mol. Cell 10 (2002) 1033–1043.
[16] E.P. Skaar, Cell Host Microbe 5 (2009) 422–424.
[17] A.K. Sia, B.E. Allred, K.N. Raymond, Curr. Opin. Chem. Biol. 17 (2013) 150–157.
[18] E.P. Skaar, PLoS Pathog. 6 (2010) e1000949.
[19] M. Caza, A. Garenaux, F. Lepine, C.M. Dozois, Mol. Microbiol. 97 (2015) 717–732.
[20] M. Valdebenito, B. Bister, R. Reissbrodt, K. Hantke, G. Winkelmann, Int. J. Med.
Microbiol. 295 (2005) 99–107.
[21] M. Valdebenito, A.L. Crumbliss, G. Winkelmann, K. Hantke, Int. J. Med. Microbiol.
296 (2006) 513–520.
Subsequent intrinsic fluorescence quenching titrations revealed that
VctP binds the Fe(III) complexes of all catecholate siderophores tested
(enterobactin, bisDHBS, 5-LICAM and Sal-5-LICAM) with higher affi-
nity than does CeuE. In addition, both CeuE and VctP display a clear
preference for tetradentate bis(catecholates), such as bisDHBS, 5-
[22] M.G. Zhu, M. Valdebenito, G. Winkelmann, K. Hantke, Microbiology 151 (2005)
2363–2372.
[23] S.I. Elshahawi, K.A. Shaaban, M.K. Kharel, J.S. Thorson, Chem. Soc. Rev. 44 (2015)
7591–7697.
83

E.J. Wilde et al.
JournalofInorganicBiochemistry190(2019)75–84
[24] B. Bister, D. Bischoff, G. Nicholson, M. Valdebenito, K. Schneider, G. Winkelmann,
K. Hantke, R. Süssmuth, Biometals 17 (2004) 471–481.
[25] M. Luo, H. Lin, M.A. Fischbach, D.R. Liu, C.T. Walsh, J.T. Groves, ACS Chem. Biol. 1
(2006) 29–32.
[40] X. Liu, Q. Du, Z. Wang, S. Liu, N. Li, Y. Chen, C. Zhu, D. Zhu, T. Wei, Y. Huang, FEBS
Lett. 586 (2012) 1240–1244.
[41] D.J. Raines, O.V. Moroz, K.S. Wilson, A.K. Duhme-Klair, Angew. Chem. Int. Ed. Eng.
52 (2013) 4595–4598.
[26] K. Hantke, G. Nicholson, W. Rabsch, G. Winkelmann, Proc. Natl. Acad. Sci. 100
(2003) 3677–3682.
[42] D.J. Raines, O.V. Moroz, E.V. Blagova, J.P. Turkenburg, K.S. Wilson, A.-K. Duhme-
Klair, Proc. Natl. Acad. Sci. 113 (2016) 5850–5855.
[27] S. Muller, M. Valdebenito, K. Hantke, Biometals 22 (2009) 691–695.
[28] A.L. Nelson, J.M. Barasch, R.M. Bunte, J.N. Weiser, Cell. Microbiol. 7 (2005)
1404–1417.
[43] E.J. Wilde, A. Hughes, E.V. Blagova, O.V. Moroz, R.P. Thomas, J.P. Turkenburg,
D.J. Raines, A.-K. Duhme-Klair, K.S. Wilson, Sci. Rep. 7 (2017) 45941–45955.
[44] H.G. Gong, M.R. Gagne, J. Am. Chem. Soc. 130 (2008) 12177–12183.
[45] X. Yu, Y. Dai, T. Yang, M.R. Gagné, H. Gong, Tetrahedron 67 (2011) 144–151.
[46] W.H. Rastetter, T.J. Erickson, M.C. Venuti, J. Organomet. Chem. 46 (1981)
3579–3590.
[29] S.M. Faruque, M.J. Albert, J.J. Mekalanos, Microbiol. Mol. Biol. Rev. 62 (1998)
1301–1314.
[30] D.A. Sack, S.R. Bradley, M.D. Sack, N. Engl. J. Med. 355 (2006) 649–651.
[31] A.R. Mey, E.E. Wyckoff, A.G. Oglesby, E. Rab, R.K. Taylor, S.M. Payne, Infect.
Immun. 70 (2002) 3419–3426.
[47] A. Chimiak, J.B. Neilands, Siderophores from Microorganisms and Plants, Springer
Berlin Heidelberg, Berlin, Heidelberg, 1984, pp. 89–96.
[32] E.E. Wyckoff, A.R. Mey, A. Leimbach, C.F. Fisher, S.M. Payne, J. Bacteriol. 188
(2006) 6515–6523.
[48] A.V. Joshua, S.K. Sharma, D.N. Abrams, Synth. Commun. 38 (2008) 434–440.
[49] A.-K. Duhme, Z. Dauter, R.C. Hider, S. Pohl, Inorg. Chem. 35 (1996) 3059–3061.
[50] F. Zhu, J. Rodriguez, T. Yang, I. Kevlishvili, E. Miller, D. Yi, S. O'Neill, M.J. Rourke,
P. Liu, M.A. Walczak, J. Am. Chem. Soc. 139 (2017) 17908–17922.
[51] L. Adak, S. Kawamura, G. Toma, T. Takenaka, K. Isozaki, H. Takaya, A. Orita,
H.C. Li, T.K.M. Shing, M. Nakamura, J. Am. Chem. Soc. 139 (2017) 10693–10701.
[52] F.M. Raymo, J.F. Stoddart, Chem. Ber. 129 (1996) 981–990.
[53] M.L. Saha, S. De, S. Pramanik, M. Schmittel, Chem. Soc. Rev. 42 (2013) 6860–6909.
[54] M. Albrecht, Chem. Rev. 101 (2001) 3457–3498.
[33] E.E. Wyckoff, S.M. Payne, Mol. Microbiol. 81 (2011) 1446–1458.
[34] E.E. Wyckoff, B.E. Allred, K.N. Raymond, S.M. Payne, J. Bacteriol. 197 (2015)
2840–2849.
[35] J. Rutz, T. Abdullah, S. Singh, V. Kalve, P. Klebba, J. Bacteriol. 173 (1991)
5964–5974.
[36] S.S. Seliger, A.R. Mey, A.M. Valle, S.M. Payne, Mol. Microbiol. 39 (2001) 801–812.
[37] S.M. Payne, A.R. Mey, E.E. Wyckoff, Microbiol. Mol. Biol. Rev. 80 (2016) 69–90.
[38] J.R. Butterton, J.A. Stoebner, S.M. Payne, S.B. Calderwood, J. Bacteriol. 174 (1992)
3729–3738.
[55] C.Y. Zamora, A.G.E. Madec, W. Neumann, E.M. Nolan, B. Imperiali, Bioorg. Med.
Chem. (2018), https://doi.org/10.1016/j.bmc.2018.04.030 in press.
[56] G.A.M. Kortman, M. Raffatellu, D.W. Swinkels, H. Tjalsma, FEMS Microbiol. Rev.
38 (2014) 1202–1234.
[39] N. Li, C. Zhang, B. Li, X. Liu, Y. Huang, S. Xu, L. Gu, J. Biol. Chem. 287 (2012)
8912–8919.
84