Endo-Hydroxamic Acid Monomers for the Assembly of a Suite of Non-native Dimeric Macrocyclic Siderophores Using Metal-Templated Synthesis

Article abstract of DOI:10.1021/acs.inorgchem.9b00878

An expedited synthesis of endo-hydroxamic acid aminocarboxylic acid (endo-HXA) compounds has been developed. These monomeric ligands are relevant to the synthesis of metal-macrocycle complexes using metal-templated synthesis (MTS), and the downstream production of apomacrocycles. Macrocycles can display useful drug properties and be used as ligands for radiometals in medical imaging applications, which supports methodological advances in accessing this class of molecule. Six endo-HXA ligands were prepared that contained methylene groups, ether atoms, or thioether atoms in different regions of the monomer (1-6). MTS using a 1:2 Fe(III)/ligand ratio furnished six dimeric hydroxamic acid macrocycles complexed with Fe(III) (1a-6a). The corresponding apomacrocycles (1b-6b) were produced upon treatment with diethylenetriaminepentaacetic acid (DTPA). Constitutional isomers of the apomacrocycles that contained one ether oxygen atom in the diamine-containing (2b) or dicarboxylic acid-containing (3b) region were well resolved by reverse-phase high-performance liquid chromatography (RP-HPLC). Density functional theory calculations were used to compute the structures and solvated molecular properties of 1b-6b and showed that the orientation of the amide bonds relative to the pseudo-C₂ axis was close to parallel in 1b, 2b, and 4b-6b but tended toward perpendicular in 3b. This conformational constraint in 3b reduced the polarity compared with 2b, consistent with the experimental trend in polarity observed using RP-HPLC. The improved synthesis of endo-HXA ligands allows expanded structural diversity in MTS-derived macrocycles and the ability to modulate macrocycle properties.

Full text of DOI:10.1021/acs.inorgchem.9b00878

Forum Article
pubs.acs.org/IC
Cite This: Inorg. Chem. XXXX, XXX, XXX−XXX
endo-Hydroxamic Acid Monomers for the Assembly of a Suite of
Non-native Dimeric Macrocyclic Siderophores Using Metal-
Templated Synthesis
Christopher J. M. Brown,† Michael P. Gotsbacher,† Jason P. Holland,‡ and Rachel Codd*,†
†School of Medical Sciences (Pharmacology), The University of Sydney, 2006 Sydney, New South Wales, Australia
‡Department of Chemistry, University of Zurich, Winterthurerstrasse 190, 8057 Zurich, Switzerland
S
* Supporting Information
ABSTRACT: An expedited synthesis of endo-hydroxamic acid aminocarboxylic acid (endo-HXA) compounds has been
developed. These monomeric ligands are relevant to the synthesis of metal−macrocycle complexes using metal-templated
synthesis (MTS), and the downstream production of apomacrocycles. Macrocycles can display useful drug properties and be
used as ligands for radiometals in medical imaging applications, which supports methodological advances in accessing this class
of molecule. Six endo-HXA ligands were prepared that contained methylene groups, ether atoms, or thioether atoms in different
regions of the monomer (1−6). MTS using a 1:2 Fe(III)/ligand ratio furnished six dimeric hydroxamic acid macrocycles
complexed with Fe(III) (1a−6a). The corresponding apomacrocycles (1b−6b) were produced upon treatment with
diethylenetriaminepentaacetic acid (DTPA). Constitutional isomers of the apomacrocycles that contained one ether oxygen
atom in the diamine-containing (2b) or dicarboxylic acid-containing (3b) region were well resolved by reverse-phase high-
performance liquid chromatography (RP-HPLC). Density functional theory calculations were used to compute the structures
and solvated molecular properties of 1b−6b and showed that the orientation of the amide bonds relative to the pseudo-C2 axis
was close to parallel in 1b, 2b, and 4b−6b but tended toward perpendicular in 3b. This conformational constraint in 3b reduced
the polarity compared with 2b, consistent with the experimental trend in polarity observed using RP-HPLC. The improved
synthesis of endo-HXA ligands allows expanded structural diversity in MTS-derived macrocycles and the ability to modulate
macrocycle properties.
macrocyclic siderophores isolated from nature include
putrebactin, avaroferrin, bisucaberin, and alcaligin (Chart 1).
These tetradentate macrocycles form 2:3 Fe(III)/ligand
complexes at pH 7 and 1:1 complexes at acidic pH values14−21
and can form complexes with other metal ions, including V(V),
Mo(VI), and Cr(V).18,21−23 Bisucaberin has been shown to
have anticancer potential through Fe(III) deprivation mech-
anisms,19 which provides the impetus to explore methods
beyond total synthesis24 to improve access to these types of
molecules and to allow increased structural diversity. This is in
INTRODUCTION
■
Almost all bacteria produce low-molecular-weight organic
ligands known as siderophores that have a high affinity toward
Fe(III).1−6 The formation of Fe(III)-siderophore complexes
occurs as the first step in siderophore-mediated bacterial Fe
acquisition, which serves to increase the availability of
insoluble Fe(III) to bacteria, essential for survival.7−10 Because
of evolutionary pressure, a vast array of siderophores are
produced with different molecular structures.4 A large subset of
siderophores contain the hydroxamic acid functional group as
the Fe(III) binding motif. Trimeric hydroxamic acid side-
rophores form hexadentate 1:1 complexes with Fe(III) and
include linear desferrioxamine B (DFOB) and macrocyclic
desferrioxamine E (DFOE).11−13 Dimeric hydroxamic acid
Special Issue: Metals in Biology: From Metallomics to Trafficking
Received: March 27, 2019
© XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.9b00878
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
diamine- or carboxylic acid-containing region of the ligand.33
In a given MTS system, the nature of the metal ion template,
the endo-HXA, and the reaction stoichiometry can direct the
architecture of the terminal macrocycle. The use of Zr(IV)
(typical coordination numbers in the range 6−8)36,37 and the
endo-HXA monomer 5-[(5-aminopentyl)(hydroxy)amino]-5-
oxopentanoic acid (PPH) in a 1:4 ratio furnished a tetrameric
hydroxamic acid macrocycle that saturated the Zr(IV)
octadentate coordination sphere.34 A selection of natural
product dimeric hydroxamic acid macrocycles have been
generated by MTS using a 1:2 Fe(III)/ligand reaction
stoichiometry.31
Chart 1. Dimeric Hydroxamic Acid Macrocyclic
Siderophores Characterized from Nature
One shortcoming of MTS in generating new macrocycles is
the availability of the endo-HXA monomers. The existing
literature syntheses38,39 provide monomers in modest yields
(about 10 mg per 5 days of synthetic undertaking), which can
hinder progress in exploring the subsequent MTS chemistry.
Our interest in increasing the traction of MTS for producing
macrocycles tailored toward a given metal ion prompted us to
reexamine the synthesis of endo-HXA monomers. In this work,
we describe an expedited synthesis of endo-HXA monomers
that gives significantly greater yields of product and uses a class
of starting reagents that allow expansion of the structural
diversity. The library of new endo-HXA monomers described
here has been used in an MTS approach to generate a set of
structurally diverse dimeric hydroxamic acid macrocycles,
which have been characterized as holo (Fe(III)-loaded) com-
plexes and apo (Fe(III)-free) ligands. These results provide the
potential to expand MTS to generate new macrocycles with
structures that deviate beyond those of natural products.
accordance with the broader potential of macrocycles as useful
drug leads and inhibitors of protein−protein interactions,25−30
which underpins developments in macrocycle synthesis.
Metal-templated synthesis (MTS) from endo-hydroxamic
acid aminocarboxylic acid (endo-HXA) monomers has been
used to furnish a range of dimeric,31 trimeric,32,33 or
tetrameric34 hydroxamic acid macrocycles. In a different
approach, trimeric macrocyclic hydroxamic acid siderophores
have been produced using Fe(III)-mediated macrolactoniza-
tion of a preassembled linear trimer.35 Trimeric DFOE was
first produced using an MTS approach from a 1:3 complex
formed in situ between Fe(III) and the endo-HXA monomer 4-
[(5-aminopentyl)(hydroxy)amino]-4-oxobutanoic acid
(PBH).32 The flanking amino and carboxylic acid groups of
contiguous PBH ligands were favorably positioned around the
Fe(III) template for ring closure using diphenylphosphoryl
azide (DPPA)-mediated peptide coupling. Ring-expanded
analogues of trimeric DFOE were subsequently generated
using MTS with the replacement of PBH with endo-HXA
monomers that contained additional methylene groups in the
a
Scheme 1. Previous Syntheses of the Archetypal endo-HXA Monomer PBH via Route a38,40,41 or b39
a
Conditions (a): (i) thiazolidine-2-thione, CH2Cl2, Et3N, 2-chloro-1-methylpyridinium iodide, rt, 24 h, yield not specified; (ii) diisobutylaluminum
hydride, toluene, −78 °C, 2 h, then −40 °C, 3 h, 61%: (iii) BnO-NH3Cl, H2O, MeOH, KOH, 0 °C, 1 h, then NaBH3CN, HCl in MeOH, rt, 3 h,
84%; (iv) succinic anhydride, pyridine, 100 °C for 1.5 h, then rt overnight, 98%; (v) H2/Raney Ni/NH3/MeOH, 0 °C, 3 h, 90%; (vi) H2, 10% Pd/
C, MeOH, HCl, rt, 3 h, 23%; (b): (i) potassium phthalimide, DMF, rt, 20 h, 84%; (ii) BnO-NH-Boc, NaH, NaI, DMF, 85 °C, 19 h, 82%; (iii)
NH2NH2, EtOH, 90 °C, 3 h; CbzCl, H2O, 1,4-dioxane, 0−20 °C, 12 h, 85% (two steps); (iv) 20% TFA, 20 °C, 2 h, 76%; (v) succinic anhydride,
t
pyridine, 100 °C, 2 h, then 20 °C for 12 h, 72%; (vi) H2, 10% Pd/C, BuOH, 1,4-dioxane, HCl, rt, 3 h, 100%.
B
DOI: 10.1021/acs.inorgchem.9b00878
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
a
Scheme 2. Improved Synthesis of endo-HXA Monomers 1−6
a
t
Conditions: (i) Boc2O, Et3N, DCM, rt, 4 h; (ii) BuO-NH-Ns (n = 1) or BnO-NH-Ns (n = 2), Ph3P, DIAD, THF, 40 °C, 8 h; (iii) 2-
mercaptoethanol, DBU, DMF, rt, 3 h; (iv) glutaric anhydride (1 and 2), 1,4-dioxane-2,6-dione (3 and 4), or 1,4-oxathiane-2,6-dione (5 and 6),
DCM, rt, 3 h; (v) 9:1 TFA/DCM, rt, 4 h (n = 1) or 1:9 TFA/DCM, rt, 2 h; H2, 10% Pd/C, tBuOH, 60 °C, 0.5 h (two steps; n = 2). PG1 was used
for 1, 2, 5, and 6. PG2 was used for 3 and 4.
efficiency, the new route allowed for expanded structural
diversity because of the availability of different types of amino
alcohol start reagents.
RESULTS AND DISCUSSION
■
Synthesis of endo-HXA Monomers. The first described
synthesis of PBH undertook a reaction between O-benzyl-N-
(4-cyanobutyl)hydroxylamine and succinic anhydride, followed
by nitrile-to-amine reduction (Raney Ni) and O-benzyl
deprotection (H2 over Pd/C).38,40,41 The reaction conditions
were harsh and furnished PBH in an overall yield of about 10%
(Scheme 1a). A subsequent synthetic scheme used milder
conditions39 and has been widely adopted by others,31,33,34,42
although the route is time-intensive and provides modest yields
of about 32% (Scheme 1b).
This work focused on PPH as the primary endo-HXA
monomer, which has been used successfully in previous MTS
approaches. PPH contains an additional methylene group in
the dicarboxylic acid-containing region than PBH and is
prepared by replacing succinic anhydride with glutaric
anhydride (PPH, 1). The replacement of 1-amino-5-pentanol
with 2-(2-aminoethoxy)ethanol gave a PPH analogue with an
ether oxygen atom in the diamine-containing region (PPHNO,
2). A thioether atom could be introduced to the diamine-
containing region using 2-[(2-aminoethyl)thio]ethanol as the
start reagent, although this was not pursued here in order to
maintain a reasonable scope of workload. The replacement of
glutaric anhydride with 1,4-dioxane-2,6-dione or 1,4-oxathiane-
2,6-dione gave PPH analogues containing an ether (PPHCO,
3) or thioether (PPHCS, 5) atom in the dicarboxylic acid-
containing region. Replacements in both the diamine- and
dicarboxylic acid-containing regions gave endo-HXA ligands
containing variable substitutions of ether and thioether groups
(PPHNOCO, 4; PPHNOCS, 6; Scheme 2).
The current work sought to improve the synthetic route
toward endo-HXA monomers at the level of yield, number of
steps, reaction times, and product diversity and are discussed
with reference to PBH. First, the more commonly used starting
reagent (Scheme 1b) 1,5-dibromopentane or 1-bromo-5-
chloropentane was replaced with 1-amino-5-pentanol (Scheme
2). The initial stage of installation of the hydroxamic acid
group involved the reaction between N-Boc-1-amino-5-
pentanol and either O-tert-butyl-N-(2-nitrophenylsulfonyl)-
hydroxylamine or O-benzyl-N-(2-nitrophenylsulfonyl)-
hydroxylamine under Mitsunobu conditions.43,44 The 2-
nitrophenylsulfonyl (nosyl) N-protecting group was selected
because it could be removed under mild conditions with a
thiolate in a fashion orthogonal to the O-(tert-butyl)-N-
alkylhydroxylamine (Scheme 2, PG1) and O-benzyl-N-
alkylhydroxylamine (Scheme 2, PG2) protecting group
systems.43−45 Following the reaction with succinic anhydride
to complete installation of the hydroxamic acid group, PBH
could be generated from the O-tert-butyl-protected intermedi-
ate upon global deprotection with trifluoroacetic acid (TFA)
or from the O-benzyl-protected intermediate with removal of
the N-Boc group with dilute TFA followed by hydrogenation
over Pd/C. The intermediates were readily purified using
autoflash chromatography, with the purity increasing from 13−
16 (80%) to 17−22 (95%) to 1−6 (95%).
Use of the O-tert-butyl protecting group for hydroxamic acid
was first pursued as a general method for endo-HXA ligands
because this could potentially enable global deprotection of the
N−OtBu and NH−Boc groups with high concentrations of
TFA. However, similar to observation from others,44 in some
instances the high concentration of TFA promoted hydrolysis
of the hydroxamic acid group. This reactivity was dependent
upon the nature of the anhydride employed in the reaction.
The N−OtBu protecting group strategy was incompatible with
1,4-dioxane-2,6-dione but was compatible with glutaric
anhydride and 1,4-oxathiane-2,6-dione. This was fortuitous in
the case of 1,4-oxathiane-2,6-dione because this mitigated
issues with sulfur poisoning of the catalyst in the Pd/C
reductive deprotection route. The 1,4-dioxane-2,6-dione
system required the use of a N−OBn protecting group,
necessitating a two-step deprotection procedure (Scheme 2,
step v). The instability of ether-containing molecules toward
N−OtBu deprotection has been reported.44 All endo-HXA
The new synthetic path gave an overall yield (Scheme 2) of
49% and an overall reaction time for five (PG1) or six (PG2)
steps of about 24 h. This compared to an overall yield of 10%
and 65 h (six steps) for the earliest devised synthesis (Scheme
1a)38,40,41 or 32% and 75 h (seven steps) for the more recent
synthesis (Scheme 1b).39 In addition to gains in the synthetic
1
1
ligands 1−6 were characterized using H and 13C, H−1H
C
DOI: 10.1021/acs.inorgchem.9b00878
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
Scheme 3. Synthesis of Dimeric Hydroxamic Acid Macrocycles as Fe(III) Complexes (Holomacrocycles 1a−6a) or as Free
Ligands (Apomacrocycles 1b−6b)
Table 1. HPLC Retention Times for Holomacrocycles 1a−6a and Apomacrocycles 1b−6b
a
b
holomacrocycle
tR (min)
m/zexp
apomacrocycle
tR (min)
m/zexp
c log P
c log P
[Fe(PP)2-MC]+
1a
2a
3a
4a
5a
6a
21.0
20.9
25.8
21.4
30.4
22.4
482.2
486.2
486.2
490.2
518.2
522.1
(PP)2-MC
1b
2b
3b
4b
5b
6b
27.6
21.4
29.7
17.9
30.2
22.9
429.1
433.4
433.3
437.2
465.2
469.3
−2.84
−5.93
−5.12
−8.21
−3.47
−6.56
0.64
−0.71
−1.48
−3.61
−0.40
−2.53
[Fe(PPNO)2-MC]+
[Fe(PPCO)2-MC]+
[Fe(PPNOCO)2-MC]+
[Fe(PPCS)2-MC]+
[Fe(PPNOCS)2-MC]+
(PPNO)2-MC
(PPCO)2-MC
(PPNOCO)2-MC
(PPCS)2-MC
(PPNOCS)2-MC
a
b
Calculated using ACD/ChemSketch 2017.2.1. Calculated using Molinspiration (https://www.molinspiration.com/cgi-bin/properties).
COSY, and H−13C HSQC NMR spectroscopy and high-
resolution mass spectrometry (HRMS; Figures S1−S6).
Synthesis of Fe(III)-Loaded Macrocycles 1a−6a Using
MTS. The set of endo-HXA monomers 1−6 allowed
examination of the synthesis of new hydroxamic acid
macrocycles using MTS. MTS relies on the formation of a
metal−ligand supramolecular assembly in which the terminal
amine and carboxylic acid groups of contiguous endo-HXA
monomers are suitably positioned to allow ring closure upon in
situ peptide coupling. This is designed to provide facile access
to metal-loaded macrocycles (holomacrocycles), with apoma-
crocycles accessible upon incubation with a high concentration
of chelator to compete for the templating metal ion (Scheme
3). The MTS reaction was conducted using a 1:2 Fe(III)/
ligand stoichiometry, with in situ amide-bond-forming
chemistry undertaken with DPPA and triethylamine.
partially resolved by the LC conditions. The experimental
MS isotope pattern at the peak maxima for 1a−6a correlated
well with the calculated isotope patterns (Figure 1, lower row).
The LC signals were ascribed to the dimeric hydroxamic acid
Fe(III) macrocycles, as determined from extraction ion
chromatogram (EIC) measurements, which correlated strongly
with the SIM traces (Figure S7, top and middle rows). EIC
signals that corresponded with the [M + H]+ adducts of the
trimeric Fe(III) macrocycles were discernible only at baseline
levels (Figure S7, lower row), which showed that the MTS
reaction stoichiometry favored formation of the dimeric
macrocycles. The signal for 5a at m/z 518.2 was present
with a signal at m/z 501.1 ascribable to the coelution of
diphenyl hydrogen phosphate [2M + H]+ as a product of
DPPA-mediated peptide coupling.46,47
Production of Apomacrocycles 1b−6b. The MTS
reaction solutions containing 1a−6a were each treated with
excess diethylenetriaminepentaacetic acid (DTPA) to compete
for the Fe(III) template and liberate the corresponding
apomacrocycles 1b−6b. The LC signals for 1b−4b and 6b
appeared as sharp, single peaks (Figure 2), with the
experimental MS isotope patterns in agreement with the
calculated data. The major MS signal correlated with the [M +
H]+ adduct of the neutral apomacrocycle, with lower-intensity
signals present for sodium and potassium adducts. The
solutions were analyzed for the presence of trimeric Fe(III)-
free macrocycles, with minor peaks detected at <1% area of the
dimeric analogues (Figure S8). In the case of 5b, two well-
resolved signals were observed at tR = 30.2 min (peak 1) and tR
= 38.5 min (peak 2). The EIC data (Figure S8, middle row)
showed that each peak was ascribable to a species with SIM
465. These two peaks were separated using semipreparative
1
The MTS reaction solution was analyzed using liquid
chromatography−mass spectrometry (LC−MS), with selected
ion monitoring (SIM) as the detection mode. The 1:2 Fe(III)/
ligand stoichiometry was selected to promote the generation of
dimeric macrocycles (Table 1). It was possible that the MTS
system might generate trimeric macrocycles because this
configuration would stabilize the octahedral coordination
preference of Fe(III). The SIM values selected for the analysis
correlated with the intrinsically charged dimeric Fe(III)
macrocycle ([M]+) and the protonated adduct ([M + H]+)
of the neutral trimeric Fe(III) macrocycle.
The signals in the LC trace were detected for each of 1a−6a
(Figure 1). Aside from the well-resolved and narrow signal for
5a, these signals were broad and commonly contained a low-
intensity shoulder, which could be due to the presence of a
distribution of Fe(III) macrocycle conformers that were
D
DOI: 10.1021/acs.inorgchem.9b00878
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
Figure 1. Complexes between Fe(III) and dimeric hydroxamic acid macrocycles (1a−6a) produced using MTS, as characterized using LC−MS
(SIM detection; first row, 1a−3a; fourth row, 4a−6a), with the experimental or calculated isotope patterns shown in each corresponding column.
HPLC for further analysis (Figure S9). As expected, MS
analysis of the separate solutions gave isotope patterns at 465.2
(peak 1) or 465.3 (peak 2). It was likely that one of these
peaks was ascribable to 5b and that the second peak was due to
an artifact of the reaction and/or side product. The
authenticity of 5b was confirmed upon the addition of an
aliquot of Fe(III) to each isolated solution. In the case of peak
1, the holomacrocycle 5a was reconstituted (Figure S9e,g),
while no LC signal was observed for 5a from a solution of
Fe(III) and peak 2 (Figure S9f). This identified peak 1 as the
apomacrocycle 5b and peak 2 as a species that had an m/z
value coincident with 5b but that was not relevant as a target.
The species in peak 2 was present in the MTS mixture
containing 5a, as detected by EIC at 465 (Figure S9a, gray).
A more accurate comparison of the LC retention times for
1b−6b was obtained from the LC trace from a mixture of
apomacrocycles that had been isolated as individual fractions
from the LC trace, with the authenticity of each compound
shown using HRMS (Figures S10−S15). The LC trace from
this composite solution (Figure 3a) showed a positive
correlation between the c log P values and the LC retention
times for 1b−6b (r = 0.82 or 0.62 for ACD/ChemSketch or
Molinspiration calculators, respectively; Figure 3b). The data
points for 2b and 3b were the most significant outliers from
the lines of best fit. Despite the presence of two ether oxygen
atoms, which would be expected to increase the polarity, 3b
eluted in a less polar window than the methylene isostere 1b.
The apomacrocycle 4b, which contained four ether oxygen
atoms, eluted in the most polar region of the LC trace. The
E
DOI: 10.1021/acs.inorgchem.9b00878
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
Figure 2. Dimeric hydroxamic acid apomacrocycles (1b−6b) detected upon removal of Fe(III) from 1a−6a, as characterized using LC−MS (SIM
detection; first row, 1b−3b; fourth row, 4b−6b), with the experimental or calculated isotope patterns shown in each corresponding column.
apomacrocycle 2b eluted in the second most polar region of
the LC trace and in a window significantly different from that
of the isomer 3b. This showed the position of the ether oxygen
atom in the diamine-containing region (2b) or the dicarboxylic
acid-containing region (3b) had a significant effect on the
solvation properties, based on the surrogate measure of the LC
retention time.
The rank order of the retention time of the holomacrocycles
(lowest to highest), 2a < 1a < 4a < 6a < 3a < 5a, differed from
that of the corresponding apomacrocycles (lowest to highest),
4b < 2b < 6b < 1b < 3b < 5b (Figure 3c). The apomacrocycle
1b moved two places higher in the rank order compared to 1a,
while the apomacrocycle 4b was ranked two places lower than
4a. At a qualitative level, these trends suggest how the
macrocycle properties might be modulated by the elements of
structural diversity, including the type and location of
backbone isosteres (methylene, oxygen, and sulfur) and the
presence or absence of Fe(III).
1H NMR Spectroscopy of the Apomacrocycle 1b. The
scale and conversion efficiency of the MTS reaction using the
standard amount of ∼4 mg of endo-HXA monomer produced
1b−6b on an analytical scale. More material of 1b was
obtained from 10 in-parallel 1:2 Fe(III)/1 MTS reaction
solutions (Table S1), which were subject to in situ ring closure
in the normal fashion. Each of the 10 reactions showed an
intense signal in LC−MS at the SIM value corresponding with
1a (SIM 482), which showed that the MTS reaction was
robust and reproducible. The signal was coincident with a
signal using visible detection at 450 nm, characteristic of
Fe(III) siderophore complexes (Figure S16). The MTS
F
DOI: 10.1021/acs.inorgchem.9b00878
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
Figure 3. LC from a mixture of isolated solutions of 1b−6b with detection using multiple SIM values (a), a plot of the c log P values of 1b−6b
from ACD/ChemSketch (black) or Molinspiration (gray) calculators versus retention time (b), and a plot of the rank order of the retention times of
1a−6a (closed circles, bottom axis) and 1b−6b (open circles, top axis) (c; using data from Figures 1 and 2).
reactions were pooled, purified, and treated with DTPA, with
molecule with C2 or Ci symmetry, with chemical shifts and
multiplicities similar to those of regionally equivalent protons
reported for putrebactin (Table 2 and Chart 2), which showed
Ci symmetry by X-ray crystallography.14
1
final purification generating about 1 mg of 1b for H NMR
1
1
spectroscopy. The signals in the H and H−1H COSY NMR
spectra (Figures 4 and S17) were consistent with 1b as a
Chart 2. Numbering System for 1b and Putrebactin
Density Functional Theory (DFT) Calculations of 1b−
6b. DFT calculations were used to calculate selected molecular
properties and estimate the solvation energetics of 1b−6b
(Table 3). Of most interest was to provide a rationale for the
significant difference in the reverse-phase high-performance
liquid chromatography (RP-HPLC) retention times observed
by experiment for the constitutional isomers 2b and 3b.
Although no symmetry constraints were imposed, all of the
structures of 1b−6b converged with pseudo-C2 symmetry with
the rotation axis positioned through the cavity of the
macrocycle (Figure 5 and Tables S2−S7).
1
Figure 4. Expansion of H−1H COSY NMR spectrum (600 MHz,
Analysis of the calculated molecular properties for 1b−6b
indicated 3b as an outlier, particularly in terms of the
DMSO-d6) of 1b.
1
Table 2. H NMR Spectroscopic Data for 1b and Putrebactin (DMSO-d6)
(PP)2-MC (1b)
putrebactin (ref 14)
a
b
a
b
assignment
δH
multiplicity
s
no.
assignment
δH
multiplicity
no.
H-1(a)
H-3(a)
H-4(a)
H-5(a)
H-7(a)
H-8(a)
H-9(a)
H-10(a)
H-11(a)
H-12(a)
9.47
2.07
1.72
2.34
7.70
3.04
1.38
1.20
1.51
3.50
2
4
4
4
2
4
4
4
4
4
H-1(a)
H-3(a)
N/A
H-4(a)
H-6(a)
H-7(a)
H-8(a)
N/A
9.71
2.26
br
t
2
4
c
t
qn
m
c
d
2.61
7.75
3.00
1.34
t
4
2
4
4
s
br
m
m
m
m
m
c
qn
H-9(a)
H-10(a)
1.48
3.47
m
t
4
4
d
m
a
b
c
d
Atom numbering as per Chart 2. Number of protons. J = 7.1 Hz. Triplet unresolved.
G
DOI: 10.1021/acs.inorgchem.9b00878
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
Table 3. Summary of the Calculated Molecular Properties and Solvation Energetics of 1b−6b
dipole
moment, μ
polarizability,
cavity
cavity surface
area (Å2)
electronic spatial
ΔHsolv
ΔSsolv
ΔGsolv
apomacrocycle
(D)
μ (au)
volume (Å3)
extent ⟨R2⟩ (au)
(kJ mol−1
)
(J K−1 mol−1
)
(kJ mol−1
)
(PP)2-MC
1b
2b
3b
4b
5b
6b
14.4
17.7
3.7
15.2
13.0
15.8
349
326
329
305
374
352
604
587
582
561
613
595
494
495
482
465
496
497
13103
13070
13040
12566
14509
14340
−41.7
−40.8
−83.1
−62.1
−87.1
−58.2
−41.7
−36.6
−83.7
−64.6
−87.5
−59.8
−41.1
−40.5
−83.0
−61.1
−86.9
−57.0
(PPNO)2-MC
(PPCO)2-MC
(PPNOCO)2-MC
(PPCS)2-MC
(PPNOCS)2-MC
Figure 5. Three projections of the optimized structures of 1b−6b calculated using the B3LYP/DGDZVP/PCM methodology.
calculated dipole moment. The calculated dipole moments (μ)
of 1b, 2b, and 4b−6b ranged between 13.0 and 17.7 D (Table
3), with the value for 3b (μ = 3.7 D) showing a reduced
polarity. Origins of this difference lie in the steric and
electronic constraints imposed by inserting the ether group
between the two carbonyl groups. In this position, the ether
oxygen atom forces the amide groups to twist and point away
from the macrocyclic cavity (essentially orthogonal to the
pseudo-C2 axis). In contrast to the other compounds in the
series, this twist in 3b reduces the polarity and dipole moment.
The effect can be seen in the electrostatic potential (ESP)
maps of 2b and 3b (Figure 6). No such difference was
observed in the calculated properties or optimized structure of
5b, which contains a thioether group in place of the ether
H
DOI: 10.1021/acs.inorgchem.9b00878
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
Figure 6. DFT-optimized structures of 2b and 3b and the calculated ESP maps (with charges in the range of −0.115e to 0.115e) showing
projections from the top face (center), as viewed down the pseudo-C2 rotation axis, and the opposing bottom face (right). ESPs are projected onto
the total electron density isosurface (at the 99.9% cutoff). Structures were optimized using the B3LYP/DGDZVP methodology, and calculations
included solvation effects in H2O using a PCM.
group of 3b (Figure S18). The larger sulfur atom in 5b allows
for increased conformational flexibility in the ring. Placing the
ether group in the center of the diaminopentane-derived region
in 2b did not induce the same conformational restrictions, with
2b retaining a high polarity. No discernible trend was observed
in the calculated solvation energies (Table 3).
conducted with 1 equiv of Fe(III) and a 1:1 ratio of two
different monomers from 1−6 could provide access to 15 new
dimeric hydroxamic acid macrocycles as a means to expand the
structural diversity.
EXPERIMENTAL SECTION
■
All reactions were performed under an inert nitrogen atmosphere.
The progress of the reactions was monitored by thin-layer
chromatography (TLC), with plates visualized using UV light at
CONCLUSION
■
This work has reported a new synthesis for endo-HXA
monomers that furnishes these ligands in higher yields (2−5
times greater) and in less time (3−4 times less) than previous
methods. The use of the nosyl protecting group adds value to
the synthesis through its orthogonality to other N-protecting
groups and its removal under mild conditions. The amino
alcohol reagent used in the first step allows an expansion of the
structural diversity in endo-HXA monomers because of the
ready availability of different analogues. The endo-HXA ligands
1−6 were used in successful MTS reactions in a 1:2 Fe(III)/
ligand stoichiometry to produce in analytical yields the
corresponding Fe(III)-loaded dimeric hydroxamic acid macro-
cycles 1a−6a, which following treatment with DTPA, gave the
apomacrocycles 1b−6b. A total of 10 in-parallel MTS reactions
using 1 gave sufficient apomacrocycle 1b for characterization
by 1H NMR spectroscopy. Apomacrocycles 1b−6b gave
different retention times on an RP-HPLC column, which
reflected different solvation properties ascribable to the
presence and position of the methylene groups and ether
and/or thioether atoms. The optimized structures and
solvation properties of the ether-containing isomers 2b and
3b were examined by DFT calculations. The results confirmed
that 3b was anomalous in the series and that an ether group
positioned between the carbonyl centers imposed steric and
electronic constraints that invoked a twist in the macrocycle,
which reduced the molecular polarity. Improved access to
endo-HXA monomers could support MTS as a more robust
method for generating metal-tailored macrocycles and other
syntheses requiring these types of ligands. Mixed-ligand MTS
254 nm or by staining with ninhydrin, vanillin, or FeCl3. H and 13C
1
NMR spectroscopy was performed on a Bruker 600 MHz AVIIII with
a TCI cryoprobe at 25 °C operating with Topspin 3.5 NMR software.
Samples were made to a concentration of 10 mg mL−1 in CDCl3
(Sigma-Aldrich, 99.8%), CD3OD (Cambridge Isotope Laboratory,
99.8%), or dimethyl sulfoxide (DMSO)-d6 (Cambridge Isotope
Laboratory, 99.9%). Chemical shifts are reported in ppm relative to
the residual solvent peaks (CDCl3, δH 7.27 ppm, δC 77.23 ppm;
CD3OD, δH 3.31 ppm, δC 49.00 ppm; DMSO-d6, δH 2.50 ppm, δC
39.52 ppm). Signal multiplicities are labeled as s (singlet), d
(doublet), t (triplet), q (quartet), qn (quintet), or m (multiplet).
Coupling constants are designated as J (Hz). Two-dimension-
al 1H−1H COSY and 1H−13C HSQC NMR spectra of 1b were
obtained using a Shigemi tube (DMS-005TB, Shigemi Inc.,
Japan). High-resolution mass spectrometry (HRMS) was conducted
on a Bruker SolariX 2XR 7T Fourier transform ion cyclotron mass
spectrometer at the School of Chemistry, University of Sydney.
General Procedures. Reagents. 1-Amino-5-pentanol (95%), 2-
(2-aminoethoxy)ethanol (98%), 1,8-diazabicyclo[5.4.0]undec-7-ene
(DBU; 98%), dichloromethane (DCM; ≥99.8%), 1,4-dioxane-2,6-
dione (common name: diglycolic anhydride, 95%), diisopropyl
azodicarboxylate (DIAD; 99%), diphenylphosphoryl azide (DPPA;
97%), di-tert-butyl dicarbonate (99%), glutaric anhydride (95%),
iron(III) acetylacetonate (97%), triethylamine (Et3N; ≥99%),
methanol (MeOH; 99.8%), 2-mercaptoethanol (≥99%), N,N-
dimethylformamide (DMF; 99.8%), succinic anhydride (≥99%),
tetrahydrofuran (THF; ≥99%), tert-butyl alcohol (≥99.5%), and
trifluoroacetic acid (TFA; 99%) were purchased from Sigma-Aldrich
(Castle Hill, Australia). Hexane and ethyl acetate were purchased
from Thermofisher Scientific (Sydney, Australia). 1,4-Oxathiane-2,6-
dione (thiodiacetic anhydride, 98%) was purchased from Fluorochem
I
DOI: 10.1021/acs.inorgchem.9b00878
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
(Hadfield, United Kingdom). All solvents were anhydrous unless
stated otherwise. All chemicals were used as received. Milli-Q H2O
was used for all experiments, as required.
(19−23 min), 3b (27−31 min), 4b (15−19 min), 5b (peak 1, 28−32
min; peak 2, 36−40 min), 6b (20−24 min). The ACN was removed
via rotary evaporation, and the fractions were then lyophilized to
dryness. The solid residue was redissolved in 200 μL of H2O, and the
solutions were pooled and analyzed by LC−MS and HRMS (Figures
S10−S15).
Instrumentation. Silica Chromatography. Flash chromatography
was performed on a Grace Reveleris X2 autoflash system using hexane
and ethyl acetate as solvents, unless specified otherwise. The cartridge
sizes and flow rates were varied depending on the scale of the
reaction: 0.1−1 g scale reaction (cartridge = 12 g; flow rate = 30 mL
min−1) and 1−8 g scale reaction (cartridge = 40 g; flow rate = 40 mL
min−1). Methods were developed using the Reveleris autogradient
feature based on Rf values from TLC experiments. Crude compounds
were dissolved in a minimum amount of ethyl acetate and injected
onto the column via the injection port.
Reverse-Phase Liquid Chromatography−Mass Spectrometry
(RP-LC−MS). RP-LC−MS was employed using an Agilent Tech-
nologies HPLC system (Santa Clara, CA) consisting of an
autoinjector (100 μL loop), an Agilent 1260 Infinity degasser, a
quaternary pump, and an Agilent 6120 series quadrupole electrospray
ionization (ESI) mass spectrometer. An Agilent C18 column reverse-
phase prepacked column (4.6 × 150 mm i.d.; particle size 5 μm) was
used for all experiments. An Agilent OpenLAB Chromatography Data
System ChemStation Edition was used to process mass chromato-
grams in each of the scans and the SIM or EIC detection modes.
Samples were made to a concentration of 1 mg mL−1 in MeOH or for
MTS samples in 1:1 DMF/H2O. The following instrumental
conditions were used: flow rate, 0.5 mL min−1; injection volume, 5
μL; spray voltage, 4 kV; capillary voltage, 3 kV; capillary temperature,
250 °C; tube lens offset, 10 V. Mobile phases were prepared by
mixing acetonitrile (ACN) formic acid (99.9:0.1, mobile phase B) and
H2O/formic acid (99.9:0.1, mobile phase A). Method A: 0−100%
ACN/H2O gradient over 40 min. Method B: 0−28% ACN/H2O
gradient over 40 min.
Semipreparative HPLC Purification of 1b. Semipreparative HPLC
was conducted on a Shimadzu LC-20 series LC system with two LC-
20AP pumps, a SIL-10AP autosampler, an SPD-20A UV/vis detector,
a FRC-10A fraction collector, and a Shimadzu ShimPack GIS-C18
column (150 × 10.0 mm i.d.; 5 mL min−1; particle size 5 μm).
Shimadzu LabSolutions Software (version 5.73) was used for data
acquisition and processing. Mobile phase A: aqueous TFA (0.05%).
Mobile phase B: ACN/TFA (99.95/0.05%).
H-Cube. Hydrogenation was performed on an H-cube Mini Plus
with a HPLC micropump. The 10% Pd/C CatCart was from
ThalesNano (ID: THS-01111). The mobile phase was prepared by
mixing tert-butyl alcohol/ethyl acetate (9:1). The method used a
temperature of 60 °C and a flow rate of 1 mL min−1 with a run time of
20 min.
General Methods. MTS. The purity of 1−6 was sufficient to
proceed to MTS without the need for resource and time-intensive
preparative HPLC. A MeOH solution (1 mL) containing 1 (4.1 mg,
17.6 μmol) was added to a MeOH solution (1 mL) containing
Fe(acac)3 (3.1 mg, 8.8 μmol) to give a 2 mg mL−1 solution with
respect to 1. The solution was stirred for 2 h, the solvent was removed
in vacuo, and the product was dried under high vacuum overnight.
DMF (4 mL) was added to the crude mixture, followed by the
addition of DPPA (29 μL, 35.2 μmol) and Et3N (10.7 μL, 35.2
μmol). The reaction mixture was stirred for 5 days, and aliquots of
100 μL were taken from the mixture and diluted with Milli-Q H2O
(100 μL) to quench the reaction. Analogous procedures were
conducted for 2−6. Reaction mixtures were analyzed by LC−MS
(positive ion) using total ion chromatogram, SIM, and EIC detection
modes.
Removing Fe(III) from Holomacrocycles 1a−6a. An aliquot of the
reaction mixture (100 μL) was taken and diluted with an equivalent
aliquot of DTPA (0.2 M), and the mixture was left to incubate at
room temperature (rt) overnight. A sample was taken and analyzed by
LC−MS.
Isolation of the Analytical Yields of Apomacrocycles 1b−6b by
HPLC. The MTS solutions treated with DTPA-containing 1b−6b
were purifed by HPLC (method B) and analyzed by HRMS. Fractions
were collected over the following elution times: 1b (26−30 min), 2b
1
In-Parallel MTS Reactions To Prepare 1b for H NMR Spectros-
copy. A total of 10 in-parallel MTS reactions were conducted using
Fe(III)/1 in a 1:2 ratio, with masses of 1 ranging between 3.7 and
15.2 mg (Table S1). After 5 days, the formation of 1a was confirmed
using LC−MS (Figure S16), and the reactions were pooled and
concentrated in vacuo. The concentrate was then diluted in 5 mL of
H2O and semipurified by a Waters C-18 SPE cartridge (2 g). The
filtrate was concentrated in vacuo and further diluted with a 0.2 M
DTPA solution (5 mL). The mixture was subjected to semi-
preparative HPLC, with a gradient of 2−40% solvent B in solvent
A from 0 to 30 min. The product eluted at 20.3 min. Fractions
containing the product were pooled and lyophilized to yield 1b (∼1
mg) characterized by NMR spectroscopy. 1H NMR (600 MHz,
DMSO-d6): δH 9.47 (s, 2H, H-1(a)), 7.70 (s, 2H, H-7(a)), 3.50 (m,
4H, H-12(a)), 3.04 (m, 4H, H-8(a)), 2.34 (m, 4H, H-5(a)), 2.07 (t, J
= 7.1 Hz, 4H, H-3(a)), 1.72 (qn, J = 7.1 Hz, 4H, H-4(a)), 1.51 (m,
4H, H-11(a)), 1.38 (qn, J = 7.1 Hz, 4H, H-9(a)), 1.20 (m, 4H, H-
10(a)) (refer to Chart 2 for atom numbering). 13C NMR (150 MHz,
DMSO-d6): δC 46.1, 37.9, 33.0, 29.9, 28.2, 25.3, 22.9, 20.4 (by
1H−13C HSQC; shifts for quaternary carbon atoms not determined).
LRMS (ESI+): m/z 429.3 ([M + H]+, 100%), 451.3 ([M + Na]+, 5%).
HRMS. Calcd for C20H36N4O6 + Na ([M + Na]+): m/z 451.25271.
Found: m/z 451.25290.
Synthesis. O-Benzyl-N-(2-nitrophenylsulfonyl)hydroxylamine.
The N,O-bis-protected hydroxylamine reagents were prepared in a
previous work43 and used in the synthesis of hydroxamic acid-based
compounds, including DFOB.44 The NH group in these reagents is
sufficiently acidic (pKa 5.1) for use in the Mitsunobu reaction. The
nosyl protecting group was used in the synthesis of linear hydroxamic
acids and the preparation of other types of ligands for radio-
chemistry.43−45 O-Benzylhydroxylamine hydrochloride (5.0 g, 31.3
mmol) was dissolved in anhydrous pyridine (60 mL), stirred
vigorously, and cooled to −5 °C. A solution of 2-nitrobenzylsulfonyl
chloride (7.1 g, 32.0 mmol) in pyridine (20 mL) was added dropwise.
The solution was left to stir at −5 °C for 30 min, then allowed to
warm to rt and stirred for a further 2 h. H2O (15 mL) was added to
terminate the reaction, and the solvents were removed in vacuo. The
residue was redissolved in EtOAc/H2O (1:1, 400 mL) and partitioned
in a separatory funnel. The organic layer was washed with 5% HCl,
H2O, and NaHCO3 (200 mL each). The organic layer was
concentrated in vacuo to yield the semipure product as an orange
crystalline solid (6.9 g, 70%). 1H NMR (600 MHz, CDCl3): δH 8.24−
8.26 (m, 1H), 8.12 (s, 1H), 7.78−7.88 (m, 3H), 7.37 (s, 5H), 5.10 (s,
2H). 13C NMR (150 MHz, CDCl3): δC 149.1, 135.6, 135.3, 132.5,
130.3, 129.7, 128.5, 124.6, 78.7. LRMS (ESI+): m/z 309.2 ([M + H]+,
100%). The data are consistent with the literature.43
O-tert-Butyl-N-(2-nitrophenylsulfonyl)hydroxylamine. O-tert-Bu-
tylhydroxylamine hydrochloride (5.06 g, 40.2 mmol) was dissolved in
anhydrous CHCl3 (80 mL) and the reaction cooled to −5 °C. Et3N
(11.7 mL, 84.4 mmol) was added dropwise, followed by the dropwise
addition of a solution of 2-nitrobenzylsulfonyl chloride (8.93, 40.2
mmol) in CHCl3 (50 mL). The reaction was stirred at −5 °C for 2 h,
warmed to rt, and left to stir overnight. The reaction mixture was
washed with H2O, 1 M HCl, H2O, 5% NaHCO3, H2O, and brine (40
mL each). The organic layer was concentrated in vacuo to yield the
1
product as a yellow crystalline solid (8.82 g, 80%). H NMR (600
MHz, CDCl3): δH 8.15−8.16 (m, 1H), 7.78−7.86 (m, 3H), 1.26 (s,
9H). 13C NMR (150 MHz, CDCl3): δC 148.5, 134.5, 133.8, 133.7,
132.7, 125.3, 82.9, 26.7. LRMS (ESI+): m/z 275.1 ([M + H]+, 100%).
The data are consistent with the literature.43
N-Boc-1-amino-5-pentanol (7). To a solution of 1-amino-5-
pentanol (1.50 g, 14.5 mmol) in CH2Cl2 (120 mL) was added a
solution of Boc2O (3.16 g, 14.5 mmol) in CH2Cl2 (10 mL). The
solution was treated with Et3N (2.22 mL, 15.9 mmol) dropwise, and
J
DOI: 10.1021/acs.inorgchem.9b00878
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
the mixture was stirred at rt for 4 h. The reaction was concentrated in
vacuo, and the crude product was purified using autoflash
chromatography (gradient: 30% EtOAc over 2 min, 30−60% over 5
min, and 60% over 2 min) to afford the product as a yellow oil (90%).
1H NMR (600 MHz, CDCl3): δH 3.67 (t, J = 6.5 Hz, 2H), 3.20 (d, J =
6.2 Hz, 2H), 1.63 (qn, J = 7.2 Hz, 2H), 1.56 (qn, J = 7.3 Hz, 2H),
1.49 (s, 9H), 1.42−1.46 (m, 2H). 13C NMR (150 MHz, CDCl3): δC
156.1, 77.1, 62.3, 40.4, 32.2, 29.8 28.4, 22.9. LRMS (ESI+): m/z 226.1
([M + Na]+, 100%). The data are consistent with the literature.48
N-Boc-2-(2-aminoethoxy)ethanol (8). To a solution of 2-(2-
aminoethoxy)ethanol (1.52 g, 14.5 mmol) in CH2Cl2 (120 mL) was
added a solution of Boc2O (3.16 g, 14.5 mmol) in CH2Cl2 (10 mL).
The solution was treated with Et3N (2.22 mL, 15.9 mmol) dropwise,
and the mixture was stirred at rt for 4 h. The reaction was
concentrated in vacuo, and the crude product was purified using
autoflash chromatography (gradient: 30% EtOAc over 2 min, 30−
60% over 5 min, and 60% over 2 min) to afford the product as a
Glutarylated N,O-Bis-protected Hydroxylamine Compounds
17−22. A further improvement was gained in the reaction step
with the anhydride (Scheme 2, step iv) by replacing pyridine with
dichloromethane and triethylamine. The triethylamine base was
subsequently determined to be unnecessary. These modifications
mitigated the high variability in yields (40−80%) and reduced the
reaction time for this step from 24 to 3 h. To a solution of 13 (401
mg, 1.46 mmol), 14 (404 mg, 1.46 mmol), 15 (450 mg, 1.46 mmol),
or 16 (453 mg, 1.46 mmol) in CH2Cl2 (10 mL) was added either
glutaric anhydride (167 mg, 1.46 mmol), 1,4-dioxane-2,6-dione (169
mg, 1.46 mmol), or 1,4-oxathiane-2,6-dione (193 mg, 1.46 mmol).
The mixture was stirred at rt for 3 h, concentrated in vacuo, and
purified using autoflash chromatography (gradient: 29% EtOAc over 2
min, 29−61% over 7 min, and 61% over 4 min) to afford the product
as a white solid (17, 21, and 22) or a clear oil (18−20; 85%).
5-[tert-Butoxy[5-[(tert-butoxycarbonyl)amino]pentyl]amino]-5-
oxopentanoic Acid (17). 1H NMR (600 MHz, CDCl3): δH 2.41 (qn,
J = 7.4 Hz, 4H), 1.93 (qn, J = 7.0 Hz, 4H), 1.60 (t, J = 7.1 Hz, 2H),
1.46−1.47 (m, 2H), 1.41 (s, 9H), 1.27 (s, 9H), 1.23 (t, J = 7.1 Hz,
4H). 13C NMR (150 MHz, CDCl3): δC 178.0, 171.3, 82.7, 79.2, 60.4,
40.4, 33.2, 32.9, 29.6, 28.4, 23.9, 21.0, 19.8, 19.6, 14.2. LRMS (ESI+):
m/z 411.3 ([M + Na]+, 100%).
1
yellow oil (90%). H NMR (600 MHz, CDCl3): δH 5.21 (s, 1H),
3.67−3.69 (m, 2H), 3.49−3.53 (m, 4H), 3.26−3.29 (m, 2H), 1.40 (s,
9H). 13C NMR (150 MHz, CDCl3): δC 156.3, 79.1, 72.2, 70.1, 61.2,
40.2, 28.4. LRMS (ESI+): m/z 228.1 ([M + Na]+, 100%). The data
are consistent with the literature.49
11-(tert-Butoxy)-2,2-dimethyl-4,12-dioxo-3,8-dioxa-5,11-diaza-
N,O-Bis-protected Hydroxylamine Compounds 13−16. To a
solution of 7 (1.50 g, 7.38 mmol) or 8 (1.51 g, 7.38 mmol) in THF
(15 mL) was added O-tert-butyl-N-(2-nitrophenylsulfonyl)-
hydroxylamine (2.03 g, 7.38 mmol) or O-benzyl-N-(2-
nitrophenylsulfonyl)hydroxylamine (2.25 g, 7.38 mmol) and PPh3
(5.80 g, 22.14 mmol). After stirring for 15 min, the reaction was
cooled to 0 °C, and DIAD (4.48 mL, 22.14 mmol) was added
dropwise. The mixture was stirred for 30 min, then heated to 40 °C,
and stirred for 8 h. The reaction was concentrated in vacuo, diluted
with diethyl ether, cooled to 0 °C, and stirred until a precipitate
formed. The mixture was filtered, and the filtrate was concentrated in
vacuo and purified using autoflash chromatography (gradient: 9%
EtOAc over 1 min, 9−28% over 10 min, and 28% over 2 min) to
furnish the semipure nosyl-protected amine compound (9−12) as an
orange oil. Products 9−12 were not further characterized and were
used directly in the next step of the synthesis. The product was
redissolved in DMF (10 mL) and treated with 2-mercaptoethanol
(517 μL, 7.38 mmol) and DBU (1.12 mL, 7.38 mmol), and the
solution was stirred at rt for 3 h. The reaction was concentrated in
vacuo, diluted with H2O, and extracted with CH2Cl2. The crude
mixture was purified using autoflash chromatography (gradient: 10%
EtOAc over 2 min, 10−25% over 4 min, and 25% over 5 min) to yield
13−16 as a yellow oil (75% over two steps).
1
hexadecan-16-oic Acid (18). H NMR (600 MHz, CDCl3): δH 3.75
(s, 2H), 3.56 (s, 4H), 3.34 (s, 2H), 2.50 (t, J = 7.0 Hz, 2H), 2.03 (t, J
= 7.0 Hz, 2H), 1.53 (s, 9H), 1.39 (s, 9H). 13C NMR (150 MHz,
CDCl3): δC 177.9, 156.2, 79.3, 69.7, 66.2, 50.5, 40.3, 33.1, 32.1, 28.4,
27.6, 19.9. LRMS (ESI+): m/z 413.2 ([M + Na]+, 100%).
11-(Benzyloxy)-2,2-dimethyl-4,12-dioxo-3,14-dioxa-5,11-diaza-
hexadecan-16-oic Acid (19). 1H NMR (600 MHz, CDCl3): δH
7.33−7.42 (m, 5H), 4.80 (s, 2H), 4.20 (s, 2H), 4.01 (s, 2H), 3.71 (s,
2H), 3.10 (s, 2H), 1.68 (qn, J = 7.3 Hz, 2H), 1.50 (qn, J = 7.4 Hz,
2H), 1.43 (s, 9H), 1.32 (s, 2H). 13C NMR (150 MHz, CDCl3): δC
172.5, 165.6, 133.7, 129.7, 129.6, 129.5, 129.0, 128.9, 76.2, 45.4, 40.3,
29.6, 28.4, 26.4, 26.2, 24.7, 22.8. LRMS (ESI+): m/z 447.2 ([M +
Na]+, 100%).
11-(Benzyloxy)-2,2-dimethyl-4,12-dioxo-3,8,14-trioxa-5,11-dia-
1
zahexadecan-16-oic Acid (20). H NMR (600 MHz, CDCl3): δH
7.39−7.45 (m, 5H), 4.91 (s, 2H), 4.31 (s, 2H), 4.20 (s, 2H), 4.10 (s,
2H), 3.66 (t, J = 5.1 Hz, 2H), 3.49 (t, J = 5.6 Hz, 2H), 3.20 (t, J = 5.5
Hz, 2H), 1.40 (s, 9H). 13C NMR (150 MHz, CDCl3): δC 172.2,
157.0, 134.5, 129.5, 128.8, 128.4, 78.7, 76.1, 69.4, 68.1, 67.6, 66.2,
60.1, 39.9, 27.4, 24.9. LRMS (ESI+): m/z 449.2 ([M + Na]+, 100%).
11-(tert-Butoxy)-2,2-dimethyl-4,12-dioxo-3-oxa-14-thia-5,11-di-
1
azahexadecan-16-oic Acid (21). H NMR (600 MHz, MeOD4): δH
3.82 (s, 2H), 3.52 (s, 2H), 3.13 (t, J = 7.0 Hz, 2H), 1.56−1.59 (m,
2H), 1.53 (s, 9H), 1.44 (s, 9H), 1.37−1.41 (m, 4H). 13C NMR (150
MHz, MeOD4): δC 172.2, 83.5, 77.9, 77.7, 77.5, 51.7, 48.8, 33.4, 33.3,
33.1, 32.9, 29.2, 27.0, 24.6. LRMS (ESI+): m/z 429.2 ([M + Na]+,
100%).
11-(tert-Butoxy)-2,2-dimethyl-4,12-dioxo-3,8-dioxa-14-thia-
5,11-diazahexadecan-16-oic Acid (22). 1H NMR (600 MHz,
CDCl3): δH 3.73 (s, 4H), 3.56 (s, 2H), 3.38 (s, 2H) 3.32 (s, 2H),
1.50 (s, 9H), 1.39 (s, 9H). 13C NMR (150 MHz, CDCl3): δC 173.2,
156.5, 89.9, 79.7, 69.8, 66.0, 50.2, 49.6, 40.3, 33.9, 33.7, 28.3, 27.6.
LRMS (ESI+): m/z 431.2 ([M + Na]+, 100%).
endo-HXA Monomers 1−6. A solution of 17 (427 mg, 1.1 mmol),
18 (429 mg, 1.1 mmol), 21 (447 mg, 1.1 mmol), or 22 (449 mg, 1.1
mmol) was prepared in CH2Cl2/TFA (1:9, 5 mL), and the solution
was stirred at rt for 4 h. The mixture was concentrated, redissolved in
toluene, and concentrated in vacuo. H2O was added, and the product
was lyophilized, yielding an orange gum. The corresponding endo-
HXA ligand 1, 2, 5, or 6 was left as a TFA salt until further reaction
was required (95%). A solution of 19 (467 mg, 1.1 mmol) or 20 (469
mg, 1.1 mmol) was prepared in tert-butyl alcohol/ethyl acetate (9:1,
18 mL) at a concentration of 50 mM and subject to hydrogenation
using a H-cube Mini Plus. The collected solution was concentrated in
vacuo, and the compound was redissolved in CH2Cl2/TFA (9:1, 5
mL) and reacted for 2 h. The mixture was concentrated, redissolved
in toluene, and concentrated in vacuo. H2O was added, and the
tert-Butyl [5-(tert-butoxyamino)pentyl]carbamate (13). 1H
NMR (600 MHz, CDCl3): δH 3.22 (d, J = 7.2 Hz, 2H), 2.94 (t, J
= 7.1 Hz, 2H), 1.58 (qn, J = 6.2 Hz, 4H), 1.54 (s, 9H), 1.43−1.47 (m,
2H), 1.28 (s, 9H). 13C NMR (150 MHz, CDCl3): δC 155.9, 79.0,
77.2, 52.8, 40.5, 30.0, 28.4, 27.0, 26.8, 24.5. LRMS (ESI+): m/z 275.3
([M + H]+, 100%).
tert-Butyl [2-[2-(tert-Butoxyamino)ethoxy]ethyl]carbamate (14).
1H NMR (600 MHz, CDCl3): δH 3.53 (t, J = 5.1 Hz, 4H), 3.48 (d, J =
4.6 Hz, 2H), 3.28 (s, 2H), 2.97 (t, J = 5.1, 2H), 1.55 (s, 9H), 1.29 (s,
9H). 13C NMR (150 MHz, CDCl3): δC 156.0, 76.8, 69.8, 67.1, 60.4,
52.5, 40.3, 28.4, 26.7. LRMS (ESI+): m/z 277.2 ([M + H]+, 100%).
tert-Butyl [5-[(Benzyloxy)amino]pentyl]carbamate (15). 1H
NMR (600 MHz, CDCl3): δH 7.26−7.32 (m, 5H), 4.67 (s, 2H),
3.07 (s, 2H), 2.89 (t, J = 7.1 Hz, 2H), 1.49 (qn, J = 7.5 Hz, 2H),
1.42−1.45 (m, 2H), 1.41 (s, 9H), 1.30 (qn, J = 7.7 Hz, 2H). 13C
NMR (150 MHz, CDCl3): δC 171.2, 137.9, 128.4, 128.3, 127.7, 77.4,
76.1, 51.9, 40.4, 29.9, 28.4, 27.0, 24.3. LRMS (ESI+): m/z 309.2 ([M
+ H]+, 100%).
tert-Butyl [2-[2-[(Benzyloxy)amino]ethoxy]ethyl]carbamate (16).
1H NMR (600 MHz, CDCl3): δH 7.26−7.33 (m, 5H), 4.69 (s, 2H),
3.58 (t, J = 5.1 Hz, 2H), 3.55 (t, J = 4.8 Hz, 2H), 3.28 (d, J = 4.6 Hz,
2H), 3.06 (t, J = 5.1 Hz, 2H), 1.42 (s, 9H). 13C NMR (150 MHz,
CDCl3): δC 156.0, 137.9, 128.4, 128.3, 127.8, 79.2, 76.0, 70.0, 67.2,
51.5, 40.4, 28.4. LRMS (ESI+): m/z 311.2 ([M + H]+, 100%).
K
DOI: 10.1021/acs.inorgchem.9b00878
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
solution was lyophilized to yield an orange gum. The corresponding
endo-HXA ligand 3 or 4 was left as a TFA salt until further reaction
was required (95%).
the potential use and application of the macrocyclic chelates described
in this work. Optimized structures were analyzed by using Chemcraft
(version 1.8, build 536b) and GaussView 6.0.16 (Gaussian Inc.,
Wallingford, MA). ESP maps (calculated in the range −0.115e to
+0.115e) were mapped onto the total electron density surface (set to
a cutoff of 99.9% density).
5-[(5-Aminopentyl)(hydroxy)amino]-5-oxopentanoic Acid (PPH)
(1). 1H NMR (600 MHz, MeOD4): δH 3.63 (t, J = 6.9 Hz, 2H, H-9),
2.92 (t, J = 7.2 Hz, 2H, H-6), 2.55 (t, J = 6.6 Hz, 2H, H-2), 2.35 (t, J
= 7.2 Hz, 2H, H-11), 1.86−1.90 (qn, J = 7.0 Hz, 2H, H-10), 1.68 (t, J
= 7.0 Hz, 4H, H-3,5), 1.38−1.40 (m, 2H, H-4). 13C NMR (150 MHz,
MeOD4): δC 175.4, 173.9, 39.2, 32.5, 30.9, 26.5, 25.7, 22.9, 19.9, 19.8.
LRMS (ESI+): m/z 233.1 ([M + H]+, 100%). HRMS: Calcd for
C10H21N2O4 ([M + H]+): m/z 233.14958. Found: m/z 233.14954.
5-[[2-(2-Aminoethoxy)ethyl](hydroxy)amino]-5-oxopentanoic
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.9b00878.
1
Acid (PPHNO, 2). H NMR (600 MHz, DMSO-d6): δH 7.81 (bs, 2H,
NH3·TFA, H-1), 3.69 (t, J = 5.9 Hz, 2H, H-6), 3.59 (t, J = 5.9 Hz, H-
3), 3.58 (m, J = 5.2 Hz, H-5), 2.95 (q, J = 5.1 Hz, 2H, H-2), 2.40 (t, J
= 7.2 Hz, 2H, H-11), 2.24 (t, J = 7.4 Hz, 2H, H-9), 1.70 (qn, J = 7.3
Hz, 2H, H-10). 13C NMR (150 MHz, DMSO-d6): δC 174.3 (C-8),
172.9 (C-12), 66.4 (C-3, -5), 46.8 (C-6), 38.6 (C-2), 33.0 (C-9), 30.9
(C-11), 20.0 (C-10). LRMS (ESI+): m/z 235.1 ([M + H]+, 100%).
HRMS: Calcd for C9H19N2O5 ([M + H]+): m/z 235.12885. Found:
m/z 235.12867.
Additional data for 1−6 and 1b, including NMR
spectroscopy and HRMS (PDF)
AUTHOR INFORMATION
Corresponding Author
*Tel.: +61 2 9351 6738. E-mail: rachel.codd@sydney.edu.au
(R.C.).
■
2-[2-[(5-Aminopentyl)(hydroxy)amino]-2-oxoethoxy]acetic Acid
(PPHCO, 3). 1H NMR (600 MHz, DMSO-d6): δH 7.75 (bs, 2H, NH3·
TFA, H-1), 4.32 (s, 2H, H-9), 4.10 (s, 2H, H-11), 3.47 (t, J = 6.8 Hz,
2H, H-6), 2.73−2.78 (m, 2H, H-2), 1.51−1.56 (m, 4H, H-3,5), 1.27
(qt, J = 7.4 Hz, 2H, H-4). 13C NMR (150 MHz, DMSO-d6): δC 171.4
(C-8), 169.3 (C-12), 67.6 (C-9), 67.2 (C-11), 46.9 (C-6), 38.7 (C-2),
26.7 (C-3, -5), 25.6 (C-4). LRMS (ESI+): m/z 235.1 ([M + H]+,
100%). HRMS: Calcd for C9H19N2O5 ([M + H]+): m/z 235.12885.
Found: m/z 235.12882.
2-[2-[[2-(2-Aminoethoxy)ethyl](hydroxy)amino]-2-oxoethoxy]-
acetic Acid (PPHNOCO, 4). 1H NMR (600 MHz, DMSO-d6): δH 7.82
(bs, 2H, NH3·TFA, H-1), 4.08 (d, J = 6.4 Hz, 4 H, H-9, -11), 3.66 (t,
J = 5.4 Hz, 2H, H-6), 3.59 (t, J = 5.8 Hz, 2H, H-5), 3.58 (t, J = 5.6 Hz,
2H, H-3) 2.95−2.96 (m, 2H, H-2). 13C NMR (150 MHz, DMSO-d6):
δC 171.2 (C-8), 169.9 (C-12), 67.5 (C-9), 67.2 (C-11), 66.3 (C-3),
66.0 (C-5), 46.9 (C-6), 38.6 (C-2). LRMS (ESI+): m/z 237.1 ([M +
H]+, 100%). HRMS: Calcd for C8H17N2O6 ([M + H]+): m/z
237.10811. Found: m/z 237.10803.
2-[[2-[(5-Aminopentyl)(hydroxy)amino]-2-oxoethyl]thio]acetic
Acid (PPHCS) (5). 1H NMR (600 MHz, DMSO-d6): δH 7.69 (bs, 2H,
NH3·TFA, H-1), 3.54 (s, 2H, H-9), 3.49 (t, J = 6.7 Hz, 2H, H-6), 3.34
(s, 2H, H-11; peak coincident with the H2O peak; analyzed by
HSQC), 2.75−2.76 (m, 2H, H-2), 1.53 (t, J = 7.1 Hz, 4H, H-3, -5),
1.27 (t, J = 7.1 Hz, 2H, H-4). 13C NMR (150 MHz, DMSO-d6): δC
170.9 (C-8), 168.7 (C-12), 47.0 (C-6), 38.7 (C-2), 33.4 (C-11), 32.6
(C-9), 26.6 (C-3, -5), 25.6 (C-4). LRMS (ESI+): m/z ([M + H]+,
100%). HRMS: Calcd for C9H19N2O4S ([M + H]+): m/z 251.10600.
Found: m/z 251.10565.
2-[[2-[[2-(2-Aminoethoxy)ethyl](hydroxy)amino]-2-oxoethyl]-
thio]acetic Acid (PPHNOCS) (6). 1H NMR (600 MHz, DMSO-d6): δH
7.76 (bs, 2H, NH3·TFA, H-1), 3.70 (t, J = 5.6 Hz, 2H, H-6), 3.56−
3.62 (m, 8H, H-3, -5, -9, -11), 2.96 (s, 2H, H-2). 13C NMR (150
MHz, DMSO-d6): δC 170.9 (C-8), 169.3 (C-12), 66.3 (C-3, -5), 46.9
(C-6), 38.6 (C-2), 32.6 (C-9, -11). LRMS (ESI+): m/z 253.1 ([M +
H]+, 100%). HRMS: Calcd for C8H17N2O5S ([M + H]+): m/z
253.08527. Found: m/z 253.08501.
ORCID
Michael P. Gotsbacher: 0000-0002-7153-1250
Jason P. Holland: 0000-0002-0066-219X
Rachel Codd: 0000-0002-2703-883X
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the Australian Research Council
(Grant DP140100092) and the Australian Government
Research Training Program Scholarship scheme (to
C.J.M.B.). A. Sresuthasan is acknowledged for useful
discussions.
REFERENCES
■
(1) Albrecht-Gary, A.-M.; Crumbliss, A. L. In Metal Ions in Biological
Systems; Sigel, A., Sigel, H., Eds.; Marcel Dekker, Inc.: New York,
1998; Vol. 35, pp 239−327.
(2) Butler, A. Iron acquisition: straight up and on the rocks? Nat.
Struct. Mol. Biol. 2003, 10, 240−241.
(3) Raymond, K. N.; Dertz, E. A. In Iron Transport in Bacteria;
Crosa, J. H., Mey, A. R., Payne, S. M., Eds.; ASM Press: Washington,
DC, 2004; pp 3−17.
(4) Hider, R. C.; Kong, X. Chemistry and biology of siderophores.
Nat. Prod. Rep. 2010, 27, 637−657.
(5) Kurth, C.; Kage, H.; Nett, M. Siderophores as molecular tools in
medical and environmental applications. Org. Biomol. Chem. 2016, 14,
8212−8227.
(6) Johnstone, T. C.; Nolan, E. M. Beyond iron: non-classical
biological functions of bacterial siderophores. Dalton Trans. 2015, 44,
6320−6339.
(7) Miethke, M. Molecular strategies of microbial iron assimilation:
from high-affinity complexes to cofactor assembly systems. Metal-
lomics 2013, 5, 15−28.
Computational Details. All calculations were conducted using
DFT, as implemented in the Gaussian16, revision A.03, suite of ab
initio quantum chemistry programs. Normal self-consistent-field and
geometry convergence criteria were employed throughout, and the
structures were optimized without using symmetry constraints. All
structures were optimized in the gas phase and in solution phase using
a polarizable continuum model (PCM) and the B3LYP/DGDZVP
methodology. Solvated phase calculations were implemented using
the SCRF keyword with default parameters and selecting H2O as the
solvent (dielectric constant, ε = 78.3553). Harmonic frequency
analysis based on analytical second derivative was used to characterize
optimized structures as local minima. The choice of the solvation
model reflects our standard aqueous phase conditions employed in
(8) Sandy, M.; Butler, A. Microbial iron acquisition: Marine and
terrestrial siderophores. Chem. Rev. 2009, 109, 4580−4595.
(9) Chu, B. C.; Garcia-Herrero, A.; Johanson, T. H.; Krewulak, K.
D.; Lau, C. K.; Peacock, R. S.; Slavinskaya, Z.; Vogel, H. J.
Siderophore uptake in bacteria and the battle for iron with the host; a
bird’s eye view. BioMetals 2010, 23, 601−611.
(10) Schalk, I. J.; Hannauer, M.; Braud, A. New roles for bacterial
siderophores in metal transport and tolerance. Environ. Microbiol.
2011, 13, 2844−2854.
(11) Codd, R.; Richardson-Sanchez, T.; Telfer, T. J.; Gotsbacher, M.
P. Advances in the chemical biology of desferrioxamine B. ACS Chem.
Biol. 2018, 13, 11−25.
L
DOI: 10.1021/acs.inorgchem.9b00878
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Forum Article
́
(12) Dhungana, S.; White, P. S.; Crumbliss, A. L. Crystal structure of
ferrioxamine B: A comparative analysis and implications for molecular
recognition. JBIC, J. Biol. Inorg. Chem. 2001, 6, 810−818.
(13) van der Helm, D.; Poling, M. The crystal structure of
ferrioxamine E. J. Am. Chem. Soc. 1976, 98, 82−86.
(32) Kachadourian, R.; Chuilon, S.; Merienne, C.; Kunesch, G.;
Deroussent, A. A new total synthesis of ferrioxamine E through metal-
templated cyclic trimerization. Supramol. Chem. 1997, 8, 301−308.
(33) Lifa, T.; Tieu, W.; Hocking, R. K.; Codd, R. Forward and
reverse (retro) iron(III)- or gallium(III)-desferrioxamine E and ring-
expanded analogs prepared using metal-templated synthesis from
endo-hydroxamic acid monomers. Inorg. Chem. 2015, 54, 3573−3583.
(34) Tieu, W.; Lifa, T.; Katsifis, A.; Codd, R. Octadentate
zirconium(IV)-loaded macrocycles with varied stoichiometry as-
sembled from hydroxamic acid monomers using metal-templated
synthesis. Inorg. Chem. 2017, 56, 3719−3728.
(14) Ledyard, K. M.; Butler, A. Structure of putrebactin, a new
dihydroxamate siderophore produced by Shewanella putrefaciens. JBIC,
J. Biol. Inorg. Chem. 1997, 2, 93−97.
̈
(15) Bottcher, T.; Clardy, J. A chimeric siderophore halts swarming.
Angew. Chem., Int. Ed. 2014, 53, 3510−3513.
(16) Soe, C. Z.; Pakchung, A. A. H.; Codd, R. Directing the
biosynthesis of putrebactin or desferrioxamine B in Shewanella
putrefaciens through the upstream inhibition of ornithine decarbox-
ylase. Chem. Biodiversity 2012, 9, 1880−1890.
(35) Wencewicz, T. A.; Oliver, A. G.; Miller, M. J. Iron(III)-
templated macrolactonization of trihydroxamate siderophores. Org.
Lett. 2012, 14, 4390−4393.
(36) Holland, J. P.; Divilov, V.; Bander, N. H.; Smith-Jones, P. M.;
Larson, S. M.; Lewis, J. S. 89Zr-DFO-J591 for immunoPET of
prostate-specific membrane antigen expression in vivo. J. Nucl. Med.
2010, 51, 1293−1300.
(17) Winkelmann, G.; Schmid, D. G.; Nicholson, G.; Jung, G.;
Colquhoun, D. J. Bisucaberin - a dihydroxamate siderophore isolated
from Vibrio salmonicida, an important pathogen of farmed Atlantic
salmon (Salmo salar). BioMetals 2002, 15, 153−160.
(18) Soe, C. Z.; Telfer, T. J.; Levina, A.; Lay, P. A.; Codd, R.
Simultaneous biosynthesis of putrebactin, avaroferrin and bisucaberin
by Shewanella putrefaciens and characterisation of complexes with
iron(III), molybdenum(VI) or chromium(V). J. Inorg. Biochem. 2016,
162, 207−215.
(19) Takahashi, A.; Nakamura, H.; Kameyama, T.; Kurasawa, S.;
Naganawa, H.; Okami, Y.; Takeuchi, T.; Umezawa, H.; Iitaka, Y.
Bisucaberin, a new siderophore, sensitizing tumor cells to macro-
phage-mediated cytolysis. II. Physico-chemical properties and
structure determination. J. Antibiot. 1987, 40, 1671−1676.
(20) Nishio, T.; Tanaka, N.; Hiratake, J.; Katsube, Y.; Ishida, Y.;
Oda, J. Isolation and structure of the novel dihydroxamate
siderophore alcaligin. J. Am. Chem. Soc. 1988, 110, 8733−8734.
(21) Codd, R.; Soe, C. Z.; Pakchung, A. A. H.; Sresutharsan, A.;
Brown, C. J. M.; Tieu, W. The chemical biology and coordinaton
chemistry of putrebactin, avaroferrin, bisucaberin, and alcaligin. JBIC,
J. Biol. Inorg. Chem. 2018, 23, 969−982.
(22) Pakchung, A. A. H.; Soe, C. Z.; Lifa, T.; Codd, R. Complexes
formed in solution between vanadium(IV)/(V) and the cyclic
dihydroxamic acid putrebactin or linear suberodihydroxamic acid.
Inorg. Chem. 2011, 50, 5978−5989.
(23) Soe, C. Z.; Pakchung, A. A. H.; Codd, R. Dinuclear
[VVO(putrebactin))2(μ-OCH3)2] formed in solution as established
from LC-MS measurements using 50V-enriched V2O5. Inorg. Chem.
2014, 53, 5852−5861.
(24) Bergeron, R. J.; McManis, J. S. The total synthesis of
bisucaberin. Tetrahedron 1989, 45, 4939−4944.
(25) Driggers, E. M.; Hale, S. P.; Lee, J.; Terrett, N. K. The
exploration of macrocycles for drug discovery - an underexploited
structural class. Nat. Rev. Drug Discovery 2008, 7, 608−624.
(26) Giordanetto, F.; Kihlberg, J. Macrocyclic drugs and clinical
candidates: What can medicinal chemists learn from their properties?
J. Med. Chem. 2014, 57, 278−295.
(27) Ardi, V. C.; Alexander, L. D.; Johnson, V. A.; McAlpine, S. R.
Macrocycles that inhibit the binding between heat shock protein 90
and TPR-containing proteins. ACS Chem. Biol. 2011, 6, 1357−1366.
(28) Marti-Centelles, V.; Pandey, M. D.; Burguete, M. I.; Luis, S. V.
Macrocyclization reactions: the importance of conformational,
configurational, and template-induced preorganisation. Chem. Rev.
2015, 115, 8736−8834.
(29) Dougherty, P. G.; Qian, Z.; Pei, D. Macrocycles as protein-
protein interaction inhibitors. Biochem. J. 2017, 474, 1109−1125.
(30) Cummings, M. D.; Sekharan, S. Structure-based macrocycle
design in small-molecule drug discovery and simple metrics to identify
opportunities for macrocyclization of small-molecule ligands. J. Med.
Chem. 2019, DOI: 10.1021/acs.jmedchem.8b01985.
(37) Holland, J. P.; Vasdev, N. Charting the mechanism and
reactivity of zirconium oxalate with hydroxamate ligands using density
functional theory: Implications in new chelate design. Dalton Trans.
2014, 43, 9872−9884.
(38) Bergeron, R. J.; Liu, Z.-R.; McManis, J. S.; Wiegand, J.
Structural alterations in desferrioxamine compatible with iron
clearance in animals. J. Med. Chem. 1992, 35, 4739−4744.
́
(39) Kadi, N.; Oves-Costales, D.; Barona-Gomez, F.; Challis, G. L. A
new family of ATP-dependent oligomerization-macrocyclization
biocatalysts. Nat. Chem. Biol. 2007, 3, 652−656.
(40) Izawa, T.; Mukaiyama, T. The partial reduction of carboxylic
acids to aldehydes via 3-acylthiazolidine-2-thiones with diisobutyla-
luminum hydride and with lithium tri-tert-butoxyaluminum hydride.
Bull. Chem. Soc. Jpn. 1979, 52, 555−558.
(41) Bergeron, R. J.; Pegram, J. J. An efficient total synthesis of
desferrioxamine B. J. Org. Chem. 1988, 53, 3131−3134.
̈
̈
(42) Rutschlin, S.; Bottcher, T. Dissecting the mechanism of
oligomerization and macrocyclization reactions of NRPS-independent
siderophore synthetases. Chem. - Eur. J. 2018, 24, 16044−16051.
(43) Reddy, P. A.; Schall, O. F.; Wheatley, J. R.; Rosik, L. O.;
McClurg, J. P.; Marshall, G. R.; Slomczynska, U. O-Protected N-(2-
nitrophenylsulfonyl)hydroxylamines: Novel reagents for the synthesis
of hydroxamates. Synthesis 2001, 2001, 1086−1092.
(44) Poreddy, A. R.; Schall, O. F.; Osiek, T. A.; Wheatley, J. R.;
Beusen, D. D.; Marshall, G. R.; Slomczynska, U. Hydroxamate-based
iron chelators: Combinatorial syntheses of desferrioxamine B
analogues and evaluation of binding affinities. J. Comb. Chem. 2004,
6, 239−254.
(45) Price, E. W.; Zeglis, B. M.; Cawthray, J. F.; Lewis, J. S.; Adam,
M. J.; Orvig, C. What a difference a carbon makes: H4octapa vs
H4C3octapa, ligands for In-111 and Lu-177 radiochemistry. Inorg.
Chem. 2014, 53, 10412−10431.
(46) Shioiri, T.; Ninomiya, K.; Yamada, S. Diphenylphosphoryl
azide. A new convenient reagent for a modified Curtus reaction and
for the peptide synthesis. J. Am. Chem. Soc. 1972, 94, 6203−6205.
(47) Montalbetti, C. A. G. N.; Falque, V. Amide bond formation and
peptide coupling. Tetrahedron 2005, 61, 10827−10852.
(48) Szabo, M.; Klein Herenbrink, C.; Christopoulos, A.; Lane, J. R.;
Capuano, B. Structure-activity relationships of privileged structures
lead to the discovery of novel biased ligands at the dopamine D2
receptor. J. Med. Chem. 2014, 57, 4924−4939.
(49) Singh, V.; Wang, S.; Kool, E. T. Genetically encoded
multispectral labeling of proteins with polyfluorophores on a DNA
backbone. J. Am. Chem. Soc. 2013, 135, 6184−6191.
(31) Sresutharsan, A.; Tieu, W.; Richardson-Sanchez, T.; Soe, C. Z.;
Codd, R. Dimeric and trimeric homo- and heteroleptic hydroxamic
acid macrocycles formed using mixed-ligand Fe(III)-based metal-
templated synthesis. J. Inorg. Biochem. 2017, 177, 344−351.
M
DOI: 10.1021/acs.inorgchem.9b00878
Inorg. Chem. XXXX, XXX, XXX−XXX