Siderophore-mediated cargo delivery to the cytoplasm of escherichia coli and pseudomonas aeruginosa: Syntheses of monofunctionalized enterobactin scaffolds and evaluation of enterobactin-cargo conjugate uptake

Article abstract of DOI:10.1021/ja3077268

The design and syntheses of monofunctionalized enterobactin (Ent, l- and d-isomers) scaffolds where one catecholate moiety of enterobactin houses an alkene, aldehyde, or carboxylic acid at the C5 position are described. These molecules are key precursors to a family of 10 enterobactin-cargo conjugates presented in this work, which were designed to probe the extent to which the Gram-negative ferric enterobactin uptake and processing machinery recognizes, transports, and utilizes derivatized enterobactin scaffolds. A series of growth recovery assays employing enterobactin-deficient E. coli ATCC 33475 (ent-) revealed that six conjugates based on l-Ent having relatively small cargos promoted E. coli growth under iron-limiting conditions whereas negligible-to-no growth recovery was observed for four conjugates with relatively large cargos. No growth recovery was observed for the enterobactin receptor-deficient strain of E. coli H1187 (fepA-) or the enterobactin esterase-deficient derivative of E. coli K-12 JW0576 (fes-), or when the d-isomer of enterobactin was employed. These results demonstrate that the E. coli ferric enterobactin transport machinery identifies and delivers select cargo-modified scaffolds to the E. coli cytoplasm. Pseudomonas aeruginosa PAO1 K648 (pvd-, pch-) exhibited greater promiscuity than that of E. coli for the uptake and utilization of the enterobactin-cargo conjugates, and growth promotion was observed for eight conjugates under iron-limiting conditions. Enterobactin may be utilized for delivering molecular cargos via its transport machinery to the cytoplasm of E. coli and P. aeruginosa thereby providing a means to overcome the Gram-negative outer membrane permeability barrier.

Full text of DOI:10.1021/ja3077268

Article
pubs.acs.org/JACS
Siderophore-Mediated Cargo Delivery to the Cytoplasm of
Escherichia coli and Pseudomonas aeruginosa: Syntheses of
Monofunctionalized Enterobactin Scaffolds and Evaluation of
Enterobactin−Cargo Conjugate Uptake
Tengfei Zheng, Justin L. Bullock, and Elizabeth M. Nolan*
Department of Chemistry, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, United States
S
* Supporting Information
ABSTRACT: The design and syntheses of monofunctionalized entero-
bactin (Ent, L- and D-isomers) scaffolds where one catecholate moiety of
enterobactin houses an alkene, aldehyde, or carboxylic acid at the C5
position are described. These molecules are key precursors to a family of
10 enterobactin−cargo conjugates presented in this work, which were
designed to probe the extent to which the Gram-negative ferric
enterobactin uptake and processing machinery recognizes, transports,
and utilizes derivatized enterobactin scaffolds. A series of growth recovery
assays employing enterobactin-deficient E. coli ATCC 33475 (ent-) revealed that six conjugates based on L-Ent having relatively
small cargos promoted E. coli growth under iron-limiting conditions whereas negligible-to-no growth recovery was observed for
four conjugates with relatively large cargos. No growth recovery was observed for the enterobactin receptor-deficient strain of E.
coli H1187 (fepA-) or the enterobactin esterase-deficient derivative of E. coli K-12 JW0576 (fes-), or when the D-isomer of
enterobactin was employed. These results demonstrate that the E. coli ferric enterobactin transport machinery identifies and
delivers select cargo-modified scaffolds to the E. coli cytoplasm. Pseudomonas aeruginosa PAO1 K648 (pvd-, pch-) exhibited
greater promiscuity than that of E. coli for the uptake and utilization of the enterobactin−cargo conjugates, and growth
promotion was observed for eight conjugates under iron-limiting conditions. Enterobactin may be utilized for delivering
molecular cargos via its transport machinery to the cytoplasm of E. coli and P. aeruginosa thereby providing a means to overcome
the Gram-negative outer membrane permeability barrier.
provide a detailed molecular and physiological understanding of
how this chelator contributes to bacterial iron homeostasis and
colonization.22 The enterobactin synthetase is comprised of
four proteins, EntBDEF, and is responsible for the production
of enterobactin from L-serine and 2,3-dihydroxybenzoic acid
(DHB).23 Following biosynthesis, Ent is exported into the
extracellular space where it scavenges Fe(III). Enterobactin
coordinates Fe(III) by its three catecholate groups with Ka ≈
1049 M−1.24 In E. coli, the outer membrane transporter FepA
(and to a lesser extent Cir and Fiu) recognizes and binds ferric
enterobactin with sub-nanomolar affinity,25,26 and provides
periplasmic entry where the siderophore forms a complex with
the periplasmic binding protein FepB.27 Subsequently, ferric
Ent is transported into the cytosol, which requires the action of
ExbBD, TonB, and FepCDG, the latter of which constitute the
inner-membrane ATP-binding cassette (ABC) transporter
system (Figure 1B).28−32 Fes, the cytosolic enterobactin
esterase, catalyzes the hydrolysis of the [Fe(Ent)]3‑ macro-
lactone,33 and the ferric reductase YgjH may subsequently assist
in iron release such that the metal ion can be used
metabolically.34 Several pathogenic Gram-negative species
INTRODUCTION
■
Siderophores are low-molecular-weight high-affinity Fe(III)
chelators that are biosynthesized and exported by bacteria,
fungi, and plants during periods of nutrient limitation for
acquiring this essential metal ion from the extracellular
milieu.1,2 Both naturally occurring and synthetic siderophore
mimics are useful for bioremediation,3 iron chelation
therapies,4,5 antibiotic drug-delivery strategies,6−14 Fe(III)
detection,15−18 protein identification,19 and pathogen cap-
ture.20,21 These types of applications benefit from or require
siderophores amenable to facile and site-specific synthetic
modification. In this work, we expand the current toolkit of site-
specifically modifiable siderophore scaffolds to include
triscatecholate enterobactin, and we report that various
synthetic enterobactin−cargo conjugates are transported to
the cytoplasm of the Gram-negative bacterial species Escherichia
coli and Pseudomonas aeruginosa by the enterobactin uptake
machinery.
Enterobactin (Ent, 1, Figure 1A) is a canonical siderophore
biosynthesized by Gram-negative species of Enterobacteriaceae
that include E. coli, Salmonella, and Klebsiella.22 Decades of
exploration pertaining to enterobactin biosynthesis and
coordination chemistry, in addition to investigations of the
proteins involved in its cellular transport and processing,
Received: August 4, 2012
Published: October 25, 2012
© 2012 American Chemical Society
18388
dx.doi.org/10.1021/ja3077268 | J. Am. Chem. Soc. 2012, 134, 18388−18400

Journal of the American Chemical Society
Article
Figure 1. Siderophores and siderophore transport machinery relevant to this work. (A) Structures of enterobactin 1 and the salmochelins MGE 2
and DGE 3. (B) Cartoon depiction of the enterobactin transport and processing machinery in E. coli.
harbor gene clusters (e.g., iroA, MccE492) responsible for post-
assembly line modifications of the enterobactin scaffold to
provide the salmochelins.33,35−38 Salmochelins are a family of
glucosylated enterobactin derivatives where the sugar moieties
are attached to the C5 position of one or more catecholate
rings (e.g., MGE 2 and DGE 3, Figure 1A).39
cargo delivery, iron and siderophore detection, and bacterial
capture.
Herein we present a family of 10 enterobactin−cargo
conjugates that are based on a monofunctionalized enterobactin
scaffold. Inspired by the salmochelins, we have derivatized
enterobactin at the C5 position of the catecholate, which
provides a point for site-specific modification without
compromising the Fe(III)-binding groups or the macrolactone
(Figure 2). Moreover, we report that the ferric enterobactin
uptake machineries of E. coli and P. aeruginosa PAO1 deliver
enterobactin-derivatized cargo to the cytoplasm of both species
under iron-deficient conditions, and that cargo size is an
important and species-specific parameter to evaluate in
enterobactin conjugate design.
Gram-negative bacteria have an outer membrane that serves
as a permeability barrier and prevents cellular entry of many
molecules, including antibiotics (e.g., vancomycin). Sidero-
phore uptake machinery provides one route to overcome this
permeability barrier,6−14 and enterobactin and its transporter
FepA have been identified as a desirable siderophore/receptor
pair for cargo delivery to Gram-negative bacterial species.13,37
FepA-mediated uptake of the ribosomal peptide antibiotics
colicin B40 and MccE492m,41 in addition to bacteriophage,42
indicates that this receptor has the capacity to transport large
molecules. Moreover, the catecholate siderophore transporters
of E. coli (e.g., Fiu, Cir) recognize synthetic catechol-modified
β-lactam antibiotics;43−46 these serendipitous observations
motivated early “Trojan horse” delivery strategies. Indeed,
small-molecule antibiotics appended to siderophore-inspired di-
and tricatecholate platforms have been evaluated for anti-
bacterial activity with mixed results.47−51 Most recently,
amoxicillin and ampicillin, β-lactam antibiotics that act in the
periplasm and target bacterial cell wall biosynthesis, were
covalently linked to a tripodal catecholate platform and
remarkably afforded ca. 102- to 103-fold enhanced activity
against P. aeruginosa PAO1 compared to the free drug.49
The ability of FepABCDG and the TonB-ExbB-ExbD system
of E. coli, as well as the enterobactin transport machinery of
other bacterial species, to recognize and provide cytosolic
transport of unnatural cargo appended to the native ligand
remains unexplored. Enterobactin exhibits C3 symmetry and
houses no unique functional group for site-specific synthetic
modification. Total syntheses of enterobactin,52−56 hydrolyti-
cally stable enterobactin analogues,57−60 and salmochelins61
have been reported. To the best of our knowledge, no
enterobactin scaffold housing a useful site-specific synthetic
handle has been presented. Such scaffolds are a prerequisite for
employing enterobactin in a variety of paradigms that include
Figure 2. Enterobactin substituted at the C5 position.
EXPERIMENTAL SECTION
■
Reagents. Dimethylformamide (DMF) and dichloromethane
(CH2Cl2) were dried over 4 Å molecular sieves or by using a VAC
solvent purification system (Vacuum Atmospheres). Anhydrous
dimethyl sulfoxide (DMSO) was purchased from Sigma-Aldrich and
used as received. The triserine lactone 4 and its D-isomer 5 were
synthesized according to a literature procedure.55 2,3-Bis(benzyloxy)-
benzoic acid 6,62 vancomycin-alkyne 7,63 and tert-butyl(2-oxo-2-(prop-
2-yn-1-ylamino)ethyl)carbamate 863 were synthesized according to
literature procedures. L-Ent 1 and its D-isomer 9 were synthesized as
reported elsewhere.55,56 tert-Butyl-3-(2-(2-(2-aminoethoxy)ethoxy)-
ethoxy)propanoate 10 was purchased from BOC Sciences (Shirley,
NY), 11-azido-3,6,9-trioxaundecan-1-amine 11 was purchased from
18389
dx.doi.org/10.1021/ja3077268 | J. Am. Chem. Soc. 2012, 134, 18388−18400

Journal of the American Chemical Society
Article
Fluka, 6-((tert-butoxycarbonyl)amino)hexanoic acid 12 was purchased
from Advanced Chem Tech, and Fmoc-PEG3-CO2H 13 was
purchased from Chem-Impex International, Inc. The syntheses of
the PEG-derivatized cargos 14−18 are provided as Supporting
Information. Methyl 5-allyl-3-methoxysalicylate 19 was obtained
from Sigma Aldrich. All other chemicals were purchased from
Sigma-Aldrich, Alfa Aesar, or TCI in the highest available purity and
used as received.
(E)-2,3-Bis(benzyloxy)-5-(prop-1-en-1-yl)benzoic Acid (22). A
30-mL portion of methanol (MeOH) was degassed with N2 for 4 h at
rt and 21 (750 mg, 2.00 mmol) was subsequently added. The mixture
was stirred at rt until 21 dissolved and PdCl2 (58 mg, 0.32 mmol) was
added to give a cloudy brown solution. The reaction was stirred at rt
for 24 h and filtered. The filtrate was concentrated and purified by
column chromatography using silica gel (1:4:5 EtOAc/hexanes/
CH2Cl2) to yield 22 as a light yellow solid (666 mg, 89%). TLC Rf =
0.4 (40% EtOAc/hexanes); mp = 140−142 °C. 1H NMR (CDCl3, 300
MHz), δ 1.88−1.90 (3H, m), 5.19 (2H, s), 5.23 (2H, s), 6.25 (1H, dq,
J = 15.9, 6.0 Hz), 6.32−6.38 (1H, m), 7.22 (1H, d, J = 2.1 Hz), 7.32−
7.51 (10H, m), 7.69 (1H, d, J = 2.1 Hz). 13C NMR (CDCl3, 125
MHz), δ 18.3, 71.4, 77.0, 115.8, 121.6, 122.7, 127.4, 127.7, 128.4,
128.7, 129.1, 129.2, 129.3, 134.7, 135.0, 135.9, 145.7, 151.3, 165.5.
HRMS (DART): [M−H]− m/z calcd, 373.1445; found, 373.1457.
N,N′-((3S,7S,11S)-11-(2,3-Bis(benzyloxy)-5-((E)-prop-1-en-1-
yl)benzamido)-2,6,10-trioxo-1,5,9-trioxacyclododecane-3,7-
diyl)bis(2,3-bis(benzyloxy)benzamide) (23). Trilactone 4 (740
mg, 2.00 mmol) and DIPEA (2.58 g, 20.0 mmol) were mixed in dry
DMSO (8 mL) and stirred for 10 min at rt to give a clear solution.
PyAOP (3.13 g, 6.07 mmol), 22 (748 mg, 2.00 mmol), and 6 (1.00 g,
2.99 mmol) were dissolved in dry DMSO (10 mL) and added to the
solution containing 4, and the reaction turned yellow and became
orange after stirring for 2 h at rt. The orange solution was mixed with
EtOAc (50 mL) and water (50 mL) and partitioned. The organic
phase was washed with brine (3 × 50 mL), dried over Na2SO4, and
concentrated to afford a yellow oil. Flash chromatography on silica gel
with a solvent gradient (10% EtOAc/hexanes to 55% EtOAc/hexanes)
yielded the product as a white foam (931 mg, 37%). TLC Rf = 0.3
General Synthetic Materials and Methods. EMD TLC silica
gel 60 F254 plates were used for analytical thin-layer chromatography.
EMD PLC silica gel 60 F254 plates of 1-mm thickness were used for
preparative TLC. Zeoprep 60HYD silica gel (40−63 μm) obtained
1
from Zeochem was used for flash chromatography. H, 19F, and 13C
NMR spectra were collected on a Varian 300 or 500 MHz
spectrophotometer, which were operated at ambient probe temper-
ature (293 K) and housed in the Department of Chemistry
Instrumentation Facility. The 1H and 13C NMR spectra were
referenced to internal standards and 19F spectra were referenced to
an external CFCl3 standard. An Avatar FTIR instrument was used to
acquire IR spectra. Optical absorption spectra were recorded on an
Agilent 8453 diode array spectrophotometer (1-cm quartz cuvettes,
Starna). General methods for high-performance liquid chromatog-
raphy and mass spectrometry, 1H and 13C NMR spectra, and IR
spectroscopic data are provided as Supporting Information.
Methyl 5-Allyl-2,3-dihydroxybenzoate (20). Methyl 5-allyl-3-
methoxysalicylate (19, 2.22 g, 10.0 mmol) and anhydrous N,N-
diisopropylethylamine (DIPEA, 1.94 g, 15.0 mmol) were dissolved in
125 mL of dry CH2Cl2 and stirred at rt for five min. The solution was
cooled to −78 °C in an acetone/dry ice bath, and boron tribromide
(BBr3, 1 M solution in CH2Cl2, 30 mL, 30 mmol) was added slowly
over ca. 10 min via a syringe to afford a yellow solution. The reaction
was stirred at −78 °C for 1 h, warmed to −30 °C over the course of 1
h, and subsequently warmed to rt and stirred for another 4.5 h. Water
(200 mL) was added slowly to quench the reaction, and the organic
phase was washed with saturated aqueous potassium carbonate
(K2CO3, 3 × 100 mL). The organic phase was dried over sodium
sulfate (Na2SO4), and the solvent was removed under reduced
pressure to afford a brown oil. Flash chromatography on silica gel with
a solvent gradient (100% hexanes to 20% EtOAc/hexanes) gave the
product as a white solid (1.09 g, 53%). TLC Rf = 0.5 (silica, CH2Cl2);
mp = 55−56 °C. 1H NMR (CDCl3, 500 MHz), δ 3.29 (2H, d, J = 7.0
Hz), 3.95 (3H, s), 5.05−5.10 (2H, m), 5.80 (1H, s), 5.91 (1H, m),
6.97 (1H, s), 7.18 (1H, s), 10.76 (1H, s). 13C NMR (CDCl3, 125
MHz), δ 39.4, 52.3, 111.9, 116.0, 119.8, 120.4, 131.1, 137.0, 144.8,
147.2, 170.7. HRMS (DART): [M+Na]+ m/z calcd, 231.0628; found,
231.0637.
1
(50% EtOAc/hexanes); mp = 100−102 °C (dec.). H NMR (CDCl3,
300 MHz), δ 1.88−1.91 (3H, m), 4.01−4.11 (3H, m), 4.16−4.22 (3H,
m), 4.91−4.98 (3H, m), 5.03−5.19 (12H, m), 6.17−6.40 (2H, m),
7.10−7.47 (32H, m), 7.66−7.71 (3H, m), 8.51−8.53 (3H, m). 13C
NMR (CDCl3, 125 MHz), δ 18.2, 40.6, 51.2, 63.9, 70.9, 76.0, 76.1,
114.2, 117.3, 120.4, 122.8, 124.1, 125.7, 126.1, 126.3, 127.4, 127.5,
127.9, 128.0, 128.2, 128.4, 128.4, 128.4, 128.7, 128.7, 129.6, 134.1,
135.8, 135.8, 136.0, 136.0, 145.5, 146.7, 151.4, 151.4, 164.7, 168.8,
168.8. HRMS (DART): [M+H]+ m/z calcd, 1250.4645; found,
1250.4653.
N,N′-((3S,7S,11S)-11-(2,3-Bis(benzyloxy)-5-formylbenz-
amido)-2,6,10-trioxo-1,5,9-trioxacyclododecane-3,7-diyl)bis-
(2,3-bis(benzyloxy)benzamide) (24). A portion of compound 23
(285 mg, 0.228 mmol) was dissolved in 1,4-dioxane (9 mL) at rt, and
water (3 mL) was added to give a colorless solution. Osmium
tetraoxide (OsO4, 68 μL of 2.5 wt % solution in 2-methyl-2-propanol,
6.7 μmol) was added and the reaction was stirred for 0.5 h at rt, which
afforded a light brown solution. Sodium periodate (NaIO4, 76.5 mg,
0.570 mmol) was then added and the reaction was stirred for another
2 h at rt. The suspension was partitioned in water (20 mL) and EtOAc
(50 mL). The organic phase was washed with 0.1 M sodium
thiosulfate (Na2S2O3, 3 × 20 mL) and brine (2 × 20 mL), and dried
over Na2SO4. Flash chromatography on silica gel with a solvent
gradient (20% EtOAc/hexanes to 65% EtOAc/hexanes) yielded the
product as white solid (165 mg, 58%). TLC Rf = 0.6 (70% EtOAc/
5-Allyl-2,3-bis(benzyloxy)benzoic Acid (21). Alkene 20 (2.18 g,
10.5 mmol), benzyl bromide (10.8 g, 60.3 mmol), and K2CO3 (24.5 g,
17.8 mmol) were combined in 200 mL of acetone at rt. The reaction
was refluxed under N2 for 16 h, which provided a yellow solution with
white solids, and the mixture was cooled to rt and filtered. The filtrate
was concentrated under reduced pressure to afford a yellow oil. The oil
was dissolved in a 375-mL mixture of 4:1 MeOH/5 M NaOH (aq).
The resulting solution was refluxed for 3.5 h and concentrated under
reduced pressure to afford a white-yellow oil. Water (300 mL) was
added to the oil, and the aqueous phase was washed with hexanes (4 ×
100 mL). The pH of the aqueous phase was adjusted to ca. 1 by
addition of 12 M HCl and the product precipitated as a white solid. A
100-mL portion of CH2Cl2 was added, and the resulting mixture was
partitioned. The aqueous phase was extracted with additional CH2Cl2
(2 × 100 mL) and the combined organic layers were dried over
Na2SO4 and concentrated under reduced pressure to yield 21 as a
white solid (3.91 g, 99%). TLC Rf = 0.55 (silica, CH2Cl2); mp = 135−
1
hexanes); mp = 74 °C (dec.). H NMR (CDCl3, 300 MHz), δ 4.03−
4.11 (3H, m), 4.18−4.26 (3H, m), 4.90−4.96 (3H, m), 5.05−5.28
(12H, m), 7.09−7.44 (31H, m), 7.65−7.67 (2H, m), 8.14−8.15 (1H,
m), 8.46−8.52 (3H, m), 9.86 (1H, s). 13C NMR (CDCl3, 125 MHz), δ
51.4, 51.4, 51.7, 64.1, 64.2, 71.0, 71.2, 76.2, 76.2, 76.5, 113.1, 117.3,
117.4, 122.9, 123.0, 124.2, 126.2, 126.3, 126.5, 127.5, 127.6, 127.8,
128.1, 128.3, 128.4, 128.4, 128.5,128.5, 128.5, 128.6, 128.8, 128.9,
132.1, 135.2, 135.3, 135.9, 135.9, 136.0, 146.7, 146.8, 151.5, 151.5,
151.7, 152.2, 163.7, 164.9, 164.9, 168.7, 168.9, 169.1, 190.6. HRMS
(DART): [M+H]+ m/z calcd, 1238.4287; found, 1238.4279.
1
136 °C. H NMR (CDCl3, 300 MHz), δ 3.38 (2H, d, J = 6.6 Hz),
5.06−5.14 (2H, m), 5.17 (2H, s), 5.22 (2H, s), 5.92 (1H, m), 7.09
(1H, d, J = 2.1 Hz), 7.31−7.50 (10H, m), 7.58 (1H, m). 13C NMR
(CDCl3, 125 MHz), δ 39.6, 71.4, 76.9, 116.7, 119.3, 122.6, 123.9,
127.8, 128.4, 128.7, 128.7, 129.1, 129.2, 134.8, 135.8, 136.2, 137.2,
145.5, 151.2, 165.6. HRMS (DART): [M−H]− m/z calcd, 373.1445;
found, 373.1439.
3,4-Bis(benzyloxy)-5-(((3S,7S,11S)-7,11-bis(2,3-bis-
(benzyloxy)benzamido)-2,6,10-trioxo-1,5,9-trioxacyclo-
dodecan-3-yl)carbamoyl)benzoic Acid (25). A portion of 24 (112
mg, 0.0903 mmol) was dissolved in 1,4-dioxane (3 mL) at rt. Sulfamic
acid (NH3SO3, 15.8 mg, 0.162 mmol) was dissolved in water (0.75
mL) and added to the dioxane solution. Sodium chlorite (NaClO2,
18390
dx.doi.org/10.1021/ja3077268 | J. Am. Chem. Soc. 2012, 134, 18388−18400

Journal of the American Chemical Society
Article
14.7 mg, 0.163 mmol) was dissolved in 0.2 mL of water and the
resulting solution was added to the reaction over the course of 10 min,
and the reaction turned yellow. After stirring for 0.5 h at rt, the
reaction was partitioned in water (10 mL) and EtOAc (20 mL), the
aqueous phase was extracted with EtOAc (2 × 10 mL), and the
combined organic phases were dried over Na2SO4. Flash chromatog-
raphy on silica gel with a solvent gradient (CH2Cl2 to 10% MeOH/
CH2Cl2) yielded the product as white solid (87 mg, 76%). TLC Rf =
0.5 (10% MeOH/CH2Cl2); mp = 128−129 °C (dec.). 1H NMR
(CDCl3, 500 MHz), δ 4.05−4.08 (3H, m), 4.22−4.25 (3H, m), 4.93−
4.98 (3H, m), 5.06−5.25 (12H, m), 7.06−7.47 (31H, m), 7.67−7.69
(2H, m), 7.86 (1H, s), 8.44−8.47 (2H, m), 8.54−8.57 (2H, m). 13C
NMR (CDCl3, 125 MHz), δ 51.4, 51.5, 51.6, 64.1, 71.1, 71.2, 76.2,
76.4, 117.5, 117.6, 123.0, 124.2, 125.4, 125.6, 126.2, 127.5, 127.6,
127.8, 128.1, 128.3, 128.4, 128.4, 128.5, 128.6, 128.7, 128.8, 128.9,
135.4, 135.6, 135.9, 136.1, 146.8, 150.7, 151.4, 151.5, 164.1, 165.0,
168.8, 168.9, 169.0, 169.3. HRMS (DART): [M+H]+ m/z calcd,
1254.4230; found, 1254.4204.
(3H, m), 5.04−5.21 (12H, m), 6.23−6.25 (1H, m), 7.09−7.45 (35H,
m), 7.64−7.66 (2H, m), 7.86 (1H, d, J = 2.0 Hz), 8.02 (1H, d, J = 2.0
Hz), 8.49−8.54 (3H, m). 13C NMR (CDCl3, 125 MHz), δ 25.6, 29.5,
38.8, 40.0, 45.3, 51.3, 51.4, 63.9, 64.1, 69.8, 69.8, 70.0, 70.3, 70.4, 70.4,
71.2, 71.2, 76.2, 76.3, 116.8, 117.5, 120.4, 123.0, 124.3, 125.4, 126.1,
126.2, 127.6, 127.6, 127.9, 128.2, 128.3, 128.4, 128.4, 128.5, 128.5,
128.6, 128.6, 128.8, 128.8, 128.9, 129.0, 130.1, 135.4, 135.7, 135.9,
136.0, 136.1, 146.8, 146.9, 149.1, 151.6, 151.8, 164.3, 164.9, 164.9,
165.8, 168.9, 169.0, 169.1, 176.2. HRMS (ESI): [M+Na]+ m/z calcd,
1560.6150; found, 1560.6269. Compound 30 was purified by
semipreparative HPLC (20% B for 5 min followed by 20−70% B
over 15 min, 4 mL/min). The product eluted at 15.1 min and was
obtained as white solid (20 mg, 58%). The analytical HPLC trace of
the purified product is reported as Supporting Information. HRMS
(ESI): [M+Na]+ m/z calcd, 1020.3333; found, 1020.3346.
N3-((3R,7R,11R)-7,11-Bis(2,3-dihydroxybenzamido)-2,6,10-
trioxo-1,5,9-trioxacyclododecan-3-yl)-N1-(1-cyclohexyl-1-oxo-
5,8,11-trioxa-2-azatridecan-13-yl)-4,5-dihydroxyisophthal-
amide (31). Compound 31 was synthesized as described for 30
except that 28 (36 mg, 29 μmol) was used instead of 25. After
purification by preparative TLC (10% MeOH/CH2Cl2), the benzyl-
protected precursor of 31 was obtained as a white oily solid (29 mg,
Enantiomers 26−28. The D-isomers of the enterobactin alkene
23, aldehyde 24, and acid 25 were synthesized as described for the L-
isomers except that triserine lactone 5 was employed instead of 4. The
synthetic procedures and characterization are provided as Supporting
Information.
1
65%). TLC Rf = 0.6 (10% MeOH/CH2Cl2). H NMR (CDCl3, 500
tert-Butyl-(1-(3-(((3S,7S,11S)-7,11-bis(2,3-dihydroxybenz-
amido)-2,6,10-trioxo-1,5,9-trioxacyclododecan-3-yl)-
carbamoyl)-4,5-dihydroxyphenyl)-1-oxo-5,8,11-trioxa-2-aza-
tridecan-13-yl)carbamate (29). Compound 25 (50 mg, 40 μmol),
PyAOP (34 mg, 60 μmol) and DIPEA (15.2 μL, 160 μmol) were
mixed in 2 mL of dry CH2Cl2 at rt. A portion of 7 (15 mg, 48 μmol)
was then added and the resulting yellow solution was stirred for 4 h at
rt. The crude reaction was washed with 0.01 N HCl (2 × 10 mL),
dried over Na2SO4, and concentrated. The benzyl-protected product
was purified by preparative TLC (10% MeOH/CH2Cl2) and obtained
as a white viscous solid (46 mg, 75%). TLC Rf = 0.7 (10% MeOH/
MHz), δ 1.17−1.25 (3H, m), 1.38−1.44 (2H, m), 1.63 (1H, m),
1.72−1.81 (4H, m), 2.01−2.06 (1H, m), 3.40−3.41 (2H, m), 3.39−
3.42 (2H, m), 3.51−3.53 (2H, m), 3.58−3.65 (12H, m), 4.01−4.06
(3H, m), 4.13−4.16 (3H, m), 4.87−4.95 (3H, m), 5.03−5.21 (12H,
m), 6.22−6.23 (1H, m), 7.09−7.45 (35H, m), 7.65−7.66 (2H, m),
7.86 (1H, s), 8.02 (1H, s), 8.49−8.54 (3H, m). 13C NMR (CDCl3, 125
MHz), δ 25.6, 29.5, 38.8, 40.0, 45.3, 51.3, 51.4, 63.9, 64.1, 69.8, 69.8,
70.0, 70.3, 70.4, 70.4, 71.2, 71.2, 76.2, 76.3, 116.8, 117.5, 120.4, 123.0,
124.3, 125.4, 126.1, 126.2, 127.6, 127.6, 127.9, 128.2, 128.3, 128.4,
128.4, 128.5, 128.5, 128.6, 128.6, 128.8, 128.8, 128.9, 129.0, 130.1,
135.4, 135.7, 135.9, 136.0, 136.1, 146.8, 146.9, 149.1, 151.6, 151.8,
164.3, 164.9, 164.9, 165.8, 168.9, 169.0, 169.1, 176.2. HRMS (ESI):
[M+Na]+ m/z calcd, 1560.6150; found, 1560.6141. Compound 31 was
purified by semipreparative HPLC (20% B for 5 min followed by 20−
70% B over 15 min, 4 mL/min). The product eluted at 14.8 min and
was obtained as white solid (5.1 mg, 27%). The analytical HPLC trace
of the purified product is reported as Supporting Information. HRMS
(ESI): [M+Na]+ m/z calcd, 1020.3333; found, 1020.3328.
1
CH2Cl2). H NMR (CDCl3, 500 MHz), δ 1.42 (9H, s), 3.27−3.28
(2H, m), 3.50−3.52 (2H, m), 3.59−3.66 (12H, m), 4.02−4.07 (3H,
m), 4.15−4.18 (3H, m), 4.90−4.94 (3H, m), 5.03−5.20 (12H, m),
7.10−7.45 (36H, m), 7.65−7.67 (2H, m), 7.85−7.85 (1H, m), 7.99
(1H, bs), 8.49−8.54 (3H, m). 13C NMR (CDCl3, 125 MHz), δ 28.3,
39.9, 40.2, 51.3, 51.4, 63.9, 64.1, 69.7, 70.0, 70.2, 70.3, 70.4, 71.1, 71.2,
76.2, 76.3, 79.0, 116.7, 117.5, 120.3, 123.0, 124.2, 125.4, 126.1, 126.2,
127.6, 127.6, 127.8, 128.2, 128.3, 128.4, 128.4, 128.4, 128.5, 128.6,
128.6, 128.7, 128.8, 128.8, 129.0, 130.2, 135.4, 135.7, 135.9, 135.9,
136.1, 146.8, 146.9, 149.0, 151.5, 151.8, 155.9, 164.2, 164.8, 164.9,
165.8, 168.9, 169.0, 169.1. HRMS (ESI): [M+Na]+ m/z calcd,
1550.5942; found, 1550.5977.
N3-((3S,7S,11S)-7,11-Bis(2,3-dihydroxybenzamido)-2,6,10-
trioxo-1,5,9-trioxacyclododecan-3-yl)-4,5-dihydroxy-N1-(1-
(naphthalen-2-yl)-1-oxo-5,8,11-trioxa-2-azatridecan-13-yl)-
isophthalamide (32). Compound 32 was synthesized as described
for 29 except that 15 (20 mg, 44 μmol) was used instead of 7. After
purification by preparative TLC (5% MeOH/CH2Cl2), the benzyl-
protected precursor of 32 was obtained as a white-yellow oily solid (37
This benzyl-protected product was dissolved in 2 mL of 1:1 EtOAc/
EtOH, the reaction flask was purged with N2, and 45 mg Pd/C (10 wt
%) was added. The reaction was stirred under H2 (1 atm) for 6 h at rt,
and the Pd/C was removed by centrifugation (13,000 rpm, 10 min).
The clear supernatant was decanted, concentrated, and re-dissolved in
a 4:2:1 mixture of 1,4-dioxane/H2O/MeOH, and purified by
semipreparative HPLC (20% B for 5 min followed by 20−70% B
over 15 min, 4 mL/min). The product eluted at 15.8 min and was
lyophilized to give 29 as white solid (15 mg, 50%). The analytical
HPLC trace of the purified product is reported as Supporting
Information. HRMS (ESI): [M+Na]+ m/z calcd, 1010.3125; found,
1010.3173.
1
mg, 59%). TLC Rf = 0.6 (10% MeOH/CH2Cl2). H NMR (CDCl3,
500 MHz), δ 3.44−3.74 (16H, m), 3.94−4.08 (4H, m), 4.12−4.16
(2H, m), 4.78−4.82 (1H, m), 4.87−4.92 (2H, m), 5.02−5.17 (12H,
m), 7.01−7.52 (39H, m), 7.58−7.59 (1H, m), 7.64−7.66 (2H, m),
7.79−7.84 (3H, m), 7.94−7.94 (1H, m), 8.29−8.31 (1H, m), 8.47−
8.50 (3H, m). 13C NMR (CDCl3, 125 MHz), δ 39.6, 39.9, 51.4, 51.4,
63.9, 64.1, 69.6, 69.7, 70.2, 70.4, 71.1, 71.2, 71.2, 76.2, 76.3, 76.3,
116.7, 117.5, 120.3, 123.1, 124.3, 124.6, 125.0, 125.2, 125.4, 126.1,
126.2, 126.2, 126.9, 127.6, 127.6, 127.9, 128.1, 128.2, 128.4, 128.4,
128.5, 128.6, 128.6, 128.8, 128.9, 128.9, 129.0, 130.0, 130.1, 130.3,
133.5, 134.5, 135.4, 135.7, 135.9, 136.0, 136.2, 146.9, 146.9, 149.0,
151.6, 151.7, 164.2, 164.9, 164.9, 165.7, 168.9, 169.0, 169.1, 169.6.
HRMS (ESI): [M+Na]+ m/z calcd, 1604.5837; found, 1604.5964.
Compound 32 was purified by semipreparative HPLC (20% B for 5
min followed by 30−55% B over 10 min, 4 mL/min) and eluted at
12.7 min. The isolated product was lyophilized and obtained as a white
solid (4.4 mg, 18%). The analytical HPLC trace of the purified product
is provided as Supporting Information. HRMS (ESI): [M+Na]+ m/z
calcd, 1064.3020; found, 1064.3084. Mass spectrometric analysis of the
crude reaction indicated M+4 in addition to the desired product 32
and suggested partial reduction of the naphthalene cargo under the
N3-((3S,7S,11S)-7,11-Bis(2,3-dihydroxybenzamido)-2,6,10-
trioxo-1,5,9-trioxacyclododecan-3-yl)-N1-(1-cyclohexyl-1-oxo-
5,8,11-trioxa-2-azatridecan-13-yl)-4,5-dihydroxyisophthal-
amide (30). Compound 30 was synthesized as described for 29
except that 14 (13.6 mg, 45.0 μmol) was used instead of 7. After
purification by preparative TLC (10% MeOH/CH2Cl2), the benzyl-
protected precursor of 30 was obtained as a white viscous solid (37
1
mg, 60%). TLC Rf = 0.6 (10% MeOH/CH2Cl2). H NMR (CDCl3,
500 MHz), δ 1.17−1.21 (3H, m), 1.37−1.43 (2H, m), 1.62−1.63 (1H,
m), 1.72−1.74 (2H, m), 1.78−1.81 (2H, m), 2.00−2.06 (1H, m),
3.39−3.42 (2H, m), 3.51−3.53 (2H, m), 3.59−3.61 (2H, m), 3.64−
3.65 (10H, m), 4.01−4.06 (3H, m), 4.13−4.17 (3H, m), 4.88−4.93
18391
dx.doi.org/10.1021/ja3077268 | J. Am. Chem. Soc. 2012, 134, 18388−18400

Journal of the American Chemical Society
Article
deprotection conditions. From analysis of HPLC peak areas, the ratio
of 32 and the partial reduction product is ca. 4:1.
MHz) δ −121.3. HRMS (ESI): [M+Na]+ m/z calcd, 1792.6434;
found, 1792.6337. Compound 35 was purified by semipreparative
HPLC (20% B for 5 min followed by 20−70% B over 10 min, 4 mL/
min) and eluted at 15.2 min. The isolated product was lyophilized and
obtained as a white solid (2.5 mg, 9%). The HPLC trace of the
purified product is provided as Supporting Information. HRMS (ESI):
[M+Na]+ m/z calcd, 1252.3617; found, 1252.3633.
N1-(1-(3-Benzylphenyl)-1-oxo-5,8,11-trioxa-2-azatridecan-
13-yl)-N3-((3S,7S,11S)-7,11-bis(2,3-dihydroxybenzamido)-
2,6,10-trioxo-1,5,9-trioxacyclododecan-3-yl)-4,5-dihydroxyiso-
phthalamide (33). Compound 33 was synthesized as described for
29 except that 16 (24 mg, 62 μmol) was used instead of 7. Partial
purification by preparative TLC (10% MeOH/CH2Cl2) afforded the
benzyl-protected precursor of 33 as a white-yellow solid with a grease
contamination (43 mg, 67%). TLC Rf = 0.6 (10% MeOH/CH2Cl2).
1H NMR (CDCl3, 500 MHz), δ 3.57−3.61 (12H, m), 3.94−3.95 (2H,
7-(4-(6-Aminohexanoyl)piperazin-1-yl)-1-cyclopropyl-6-flu-
oro-4-oxo-1,4-dihydroquinoline-3-carboxylic Acid (36). Cipro-
floxacin (37, 331 mg, 1.00 mmol) and DIPEA (1.0 mL, 5.7 mmol)
were mixed in 6 mL of dry CH2Cl2, and TMSCl (370 μL, 2.91 mmol)
was added to give a clear yellow solution. 6-((tert-Butoxycarbonyl)-
amino)hexanoic acid (12, 346 mg, 1.50 mmol), PyAOP (834 mg, 1.60
mmol), and DIPEA (700 μL, 4.02 mmol) were dissolved in 4 mL of
dry CH2Cl2, and the two solutions were combined and stirred
overnight at rt. The reaction was quenched with MeOH (10 mL), and
the resulting solution was concentrated to dryness, and the crude
product was redissolved in 40 mL of EtOAc. The organic phase was
washed with 10 mM HCl (2 × 40 mL) and saturated aqueous
NaHCO3 (2 × 40 mL), dried over Na2SO4, and purified by flash
chromatography on silica gel (3% MeOH/CH2Cl2) to give 38 as
d, J = 6.0) 3.97−4.05 (3H, m), 4.07−4.15 (3H, m), 4.85−4.90 (3H,
m), 5.01−5.17 (12H, m), 7.01−7.40 (30H, m) 7.62−7.70 (3H, m),
7.82 (1H, d, J = 2.0), 7.99−8.00 (1H, d, J = 2.0), 8.47−8.51 (3H, m)
HRMS (ESI): [M+Na]+ m/z calcd, 1644.6150; found, 1644.6105. A
portion (32.5 mg, 20.0 μmol) of this material was carried on without
further purification or characterization. Compound 33 was purified by
semipreparative HPLC (20% B for 5 min followed by 20−70% B over
15 min, 4 mL/min). The product eluted at 15.8 min and was obtained
as white solid (13.5 mg, 62%). The analytical HPLC trace of the
purified product is provided as Supporting Information. HRMS (ESI):
[M+Na]+ m/z calcd, 1104.3333; found, 1104.3305.
1
yellow solid (243 mg, 45%). TLC Rf = 0.7 (5% MeOH/CH2Cl2). H
N3-((3S,7S,11S)-7,11-Bis(2,3-dihydroxybenzamido)-2,6,10-
trioxo-1,5,9-trioxacyclododecan-3-yl)-4,5-dihydroxy-N1-(1-
oxo-1-(11-oxo-2,3,5,6,7,11-hexahydro-1H-pyrano[2,3-f ]-
pyrido[3,2,1-ij]quinolin-10-yl)-5,8,11-trioxa-2-azatridecan-13-
yl)isophthalamide (34). Compound 34 was synthesized as described
for 29 except that 17 (18 mg, 39 μmol) was used instead of 7. After
purification by preparative TLC (10% MeOH/CH2Cl2) the benzyl-
protected precursor of 34 was obtained as an orange oily solid (18 mg,
NMR (CDCl3, 300 MHz), δ 1.14−1.20 (2H, m), 1.32−1.53 (13H, m),
1.59−1.69 (2H, m), 2.36 (2H, t, J = 6.0 Hz), 3.08 (2H, dt, J = 6.3, 6.3
Hz), 3.26−3.56 (4H, m), 3.51−3.59 (1H, m), 3.69−3.82 (4H, m),
4.68 (1H, bs), 7.32 (1H, d, J = 7.2 Hz), 7.82 (1H, d, J = 12.9 Hz), 8.60
(1H, s), 14.9 (1H, bs). 13C NMR (CDCl3, 125 MHz), δ 8.1, 24.7, 26.4,
28.3, 29.8, 32.9, 35.3, 40.2, 41.0, 45.1, 49.3, 49.9, 78.9, 105.0, 107.7,
111.9, 112.1, 119.6, 119.7, 138.8, 145.2, 145.3, 147.3, 152.4, 154.4,
155.9, 166.6, 171.4, 176.7. 19F NMR (CDCl3, 282 MHz), δ −121.1.
HRMS (ESI): [M+H]+ m/z calcd, 545.2775; found, 545.2768.
The TFA salt of 36 (202 mg, 98%) was obtained as a yellow solid
from 38 (201 mg, 0.369 mmol) by stirring 38 in 40% TFA/CH2Cl2 at
rt for 3 h and removing the solvent. TLC Rf = 0.1 (10% MeOH/
CH2Cl2). 1H NMR (CD3OD, 300 MHz), δ 1.41−1.52 (4H, m), 1.65−
1.77 (4H, m), 2.52 (2H, t, J = 7.2 Hz), 2.96 (2H, t, J = 7.2 Hz), 3.34−
3.43 (4H, m), 3.82 (5H, m), 7.57 (1H, d, J = 7.5 Hz), 7.85 (1H, d, J =
13.2 Hz), 8.76 (1H, s). 13C NMR (CDCl3, 125 MHz), δ 7.8, 23.8,
25.4, 26.5, 26.6, 32.2, 35.4, 39.0, 39.1, 41.0, 45.0, 48.1, 48.3, 48.5, 48.6,
48.8, 49.0, 49.1, 49.5, 105.0, 107.0, 111.6, 111.8, 119.3, 119.4, 138.8,
145.1, 145.2, 147.4, 152.3, 154.3, 167.3, 171.8, 176.5. 19F NMR
(CDCl3, 282 MHz), δ −76.0, −120.9. HRMS (ESI): [M+H]+ m/z
calcd, 445.2251; found, 445.2255.
1
26%). TLC Rf = 0.7 (10% MeOH/CH2Cl2). H NMR (CDCl3, 500
MHz), δ 1.93−1.95 (4H, m), 2.71−2.83 (4H, m), 3.26−3.32 (4H, m),
3.56−3.69 (16H, m), 3.99−4.18 (6H, m), 4.88−4.94 (3H, m), 5.01−
5.18 (12H, m), 6.94 (1H, s), 7.06−7.43 (35H, m), 7.62−7.66 (2H, m),
7.80−7.80 (1H, m), 7.97−7.97 (1H, m), 8.47−8.53 (4H, m), 9.02−
9.03 (1H, m). 13C NMR (CDCl3, 125 MHz), δ 19.9, 20.0, 21.0, 27.3,
39.4, 40.1, 49.7, 50.2, 51.5, 64.1, 69.9, 71.1, 71.2, 76.3, 105.4, 108.1,
115.9, 117.5, 119.8, 123.0, 124.3, 125.7, 126.3, 127.2, 127.6, 127.6,
127.8, 128.1, 128.2, 128.5, 128.5, 128.6, 128.9, 128.9, 129.0, 130.0,
135.7, 136.0, 136.2, 146.9, 148.2, 148.3, 149.0, 151.6, 151.7, 152.6,
162.9, 164.4, 165.0, 165.0, 168.9, 169.1. HRMS (ESI): [M+Na]+ m/z
calcd, 1717.6313; found, 1717.6287. Compound 34 was purified by
semipreparative HPLC (20% B for 5 min followed by 20−70% B over
15 min, 4 mL/min). The product eluted at 17.1 min and was obtained
as an orange solid (4.5 mg, 48%). The analytical HPLC trace of the
purified product is provided as Supporting Information. HRMS (ESI):
[M+Na]+ m/z calcd, 1177.3496; found, 1177.3540.
7-(4-(1-(3-(((3S,7S,11S)-7,11-Bis(2,3-dihydroxybenzamido)-
2,6,10-trioxo-1,5,9-trioxacyclododecan-3-yl)carbamoyl)-4,5-di-
hydroxyphenyl)-1-oxo-5,8,11-trioxa-2-azatetradecan-14-oyl)-
piperazin-1-yl)-1-cyclopropyl-6-fluoro-4-oxo-1,4-dihydro-
quinoline-3-carboxylic Acid (35). Compound 35 was synthesized
as described for 29 except that 18 (26 mg, 48 μmol) was used instead
of 7, and TMSCl (10 μL, 79 μmol) and DIPEA (15 μL, 0.16 mmol)
were mixed with 18 before addition to the solution containing 25.
After purification by preparative TLC (10% MeOH/CH2Cl2), the
benzyl-protected precursor of 35 was obtained as a yellow oily solid
(46 mg, 65%). TLC Rf = 0.65 (10% MeOH/CH2Cl2). 1H NMR
(CDCl3, 500 MHz), δ 1.13 (2H, bs), 1.33 (2H, bs), 2.64 (2H, bs),
3.23−3.30 (4H, m), 3.51 (1H, bs), 3.63 (14H, bs), 3.79 (4H, bs),
3.99−4.04 (3H, m), 4.11−4.14 (3H, m), 4.86−4.91 (3H, m), 5.01−
5.19 (12H, m), 7.06−7.43 (39H, m), 7.59−7.61 (2H, m), 7.83 (1H, s),
7.97−7.99 (2H, m), 8.45−8.49 (3H, m), 8.69 (1H, s). 13C NMR
(CDCl3, 125 MHz), δ 8.2, 33.4, 35.4, 40.0, 41.1, 45.3, 49.3, 50.0, 51.3,
51.4, 51.4, 63.9, 64.1, 67.1, 69.7, 70.2, 70.3, 70.4, 70.5, 71.2, 71.3, 76.2,
76.3, 105.2, 108.0, 112.3, 112.4, 116.7, 117.5, 120.0, 120.0, 120.5,
123.0, 124.3, 125.6, 126.1, 126.1, 127.6, 127.6, 127.8, 128.2, 128.3,
128.4, 128.4, 128.5, 128.6, 128.6, 128.8, 128.8, 128.8, 129.0, 130.2,
135.5, 135.7, 135.9, 136.0, 136.1, 138.9, 145.2, 145.3, 146.8, 146.8,
147.4, 149.0, 151.6, 151.6, 151.8, 152.4, 154.4, 164.2, 164.9, 164.9,
165.8, 166.9, 168.9, 169.0, 169.1, 169.7, 176.9. 19F NMR (CDCl3, 282
7-(4-(6-(3-(((3S,7S,11S)-7,11-Bis(2,3-dihydroxybenzamido)-
2,6,10-trioxo-1,5,9-trioxacyclododecan-3-yl)carbamoyl)-4,5-
dihydroxybenzamido)hexanoyl)piperazin-1-yl)-1-cyclopropyl-
6-fluoro-4-oxo-1,4-dihydroquinoline-3-carboxylic Acid (40).
Compound 40 was synthesized as described for 35 except that
DMSO (1.5 mL) was used as the solvent and compound 36 (19.4 mg,
34.8 μmol) was used instead of 18. After preparative TLC purification
(10% MeOH/CH2Cl2), 39 was obtained as white viscous solid (17
1
mg, 74%). TLC Rf = 0.6 (10% MeOH/CH2Cl2). H NMR (CDCl3,
300 MHz), δ 1.17−1.83 (10H, m), 2.40 (2H, bs), 3.29−3.44 (6H, m),
3.70−3.86 (5H, m), 4.02−4.17 (6H, m), 4.86−4.93 (3H, m), 5.04−
5.21 (12H, m), 7.07−7.42 (33H, m), 7.60−7.64 (2H, m), 7.85−8.05
(3H, m), 8.47−8.50 (3H, m), 8.74 (1H, bs), 15.0 (1H, bs).13C NMR
(CDCl3, 125 MHz), δ 8.1, 12.3, 17.2, 18.6, 24.4, 26.3, 26.4, 26.5, 29.0,
32.8, 34.7, 39.7, 41.2, 45.3, 46.2, 46.3, 51.4, 51.5, 51.5, 52.0, 54.8, 63.9,
64.1, 64.2, 71.1, 71.2, 71.2, 76.2, 76.3, 76.3, 105.2, 109.5, 113.0, 113.2,
116.6, 117.5, 120.1, 123.0, 124.3, 125.5, 126.1, 127.6, 127.8, 128.2,
128.3, 128.4, 128.4, 128.5, 128.6, 128.6, 128.6, 128.8, 128.9, 128.9,
129.0, 130.3, 135.5, 135.8, 135.9, 136.0, 136.1, 138.1, 145.4, 146.8,
148.4, 149.0, 151.6, 151.8, 152.3, 164.4, 164.9, 165.0, 165.8, 166.1,
168.8, 169.0, 169.1, 171.5. HRMS (ESI): [M+H]+ m/z calcd,
1680.6303; found, 1680.6352. Compound 40 was purified by
semipreparative HPLC (20% B for 5 min followed by 20−70% B
over 15 min, 4 mL/min) and eluted at 16.1 min. The isolated product
was lyophilized and obtained as a white-yellow solid (6.7 mg, 59%).
The HPLC trace of the purified product is provided as Supporting
18392
dx.doi.org/10.1021/ja3077268 | J. Am. Chem. Soc. 2012, 134, 18388−18400

Journal of the American Chemical Society
Article
Scheme 1. Syntheses of 22 (Top) and Monofunctionalized Enterobactin Scaffolds (Bottom)
Information. HRMS (ESI): [M+H]+ m/z calcd, 1140.3486; found,
1140.3482.
tert-Butyl(2-(((1-(1-(3-(((3S,7S,11S)-7,11-bis(2,3-dihydroxy-
benzamido)-2,6,10-trioxo-1,5,9-trioxacyclododecan-3-yl)-
carbamoyl)-4,5-dihydroxyphenyl)-1-oxo-5,8,11-trioxa-2-aza-
tridecan-13-yl)-1H-1,2,3-triazol-4-yl)methyl)amino)-2-
oxoethyl)carbamate (43). Compound 43 was synthesized as
described for 42 except that a DMSO solution of 7 (2.8 mg/mL, 13
mM, 25 μL) was used instead of 8. HPLC purification gave 3.3 mg of
the benzyl-protected precursor of 43 as a white solid (58%). HRMS
(ESI): [M+H]+ m/z calcd, 1688.6489; found, 1688.6421. Compound
43 (3.3 mg, 33%) was obtained from the precursor (13.3 mg, 7.88
μmol; obtained from four 25-μL scale click reactions) following the
same procedure as synthesizing 42. HPLC purification (0% B for 5
min followed by 0−45% B over 8 min, 4 mL/min) afforded 43 as a
white solid with a retention time of 12.8 min. The HPLC trace of the
purified product is reported as Supporting Information. HRMS (ESI):
[M+H]+ m/z calcd, 1126.3853; found, 1126.3832.
Growth Recovery Assays. General microbiology methods are
included as Supporting Information. Overnight cultures were prepared
by inoculating 5 mL of LB (E. coli) or LB base supplemented with 2.5
g/L NaCl (P. aeruginosa) with the appropriate freezer stocks and the
cultures were incubated at 37 °C in a tabletop incubator shaker set at
150 rpm. The overnight culture was diluted 1:100 into 5 mL of fresh
media with or without 200 μM 2,2′-dipyridyl (DP) and incubated at
37 °C with shaking at 150 rpm until the optical density at 600 nm
(OD600) reached 0.6. The cultures were diluted to an OD600 value of
0.001 in 50% reduced MHB medium (10.5 g/L) with or without 200
μM (E. coli) or 600 μM DP (P. aeruginosa). A 90-μL aliquot of the
diluted culture was mixed with a 10-μL aliquot of a 10x solution of the
siderophore or siderophore−cargo conjugate in a 96-well plate, which
was wrapped in parafilm and incubated at 30 °C with shaking at 150
rpm for 19 h. Bacterial growth was assayed by measuring OD600 using
a BioTek Synergy HT plate reader. Each well condition was prepared
in duplicate, and three independent replicates of each assay were
conducted on different days. The resulting mean OD600 are reported,
and the error bars are the standard deviation of the mean obtained
from the three independent replicates.
N1-(2-(2-(2-(2-Azidoethoxy)ethoxy)ethoxy)ethyl)-N3-
((3S,7S,11S)-7,11-bis(2,3-dihydroxybenzamido)-2,6,10-trioxo-
1,5,9-trioxacyclododecan-3-yl)-4,5-dihydroxyisophthalamide
(41). 11-Azido-3,6,9-trioxaundecan-1-amine (11, 8.2 μL, 42 μmol) and
25 (40 mg, 32 μmol) were dissolved in 1 mL of dry CH2Cl2. PyAOP
(33.2 mg, 63.8 μmol) and DIPEA (22.2 μL, 128 μmol) were added to
give a light yellow solution. The reaction was stirred for 4 h at rt and
concentrated, and the crude product was purified by preparative TLC
(50% EtOAc/CH2Cl2) to afford 41 as a light yellow oil (31 mg, 68%).
TLC Rf = 0.3 (50% EtOAc/CH2Cl2). 1H NMR (CDCl3, 300 MHz), δ
3.34 (2H, t, J = 5.1 Hz), 3.61−3.69 (14H, m), 3.97−4.18 (6H, m),
4.88−4.94 (3H, m), 5.02−5.22 (12H, m), 7.08−7.46 (34H, m), 7.64−
7.67 (2H, m), 7.85 (1H, d, J = 1.8 Hz), 7.99 (1H, d, J = 2.1 Hz), 8.48−
8.52 (3H, m). 13C NMR (CDCl3, 125 MHz), δ 40.0, 50.6, 51.4, 64.0,
64.1, 69.7, 69.9, 70.3, 70.6, 71.2, 71.2, 76.3, 116.7, 117.5, 120.4, 123.1,
124.3, 125.5, 126.2, 126.2, 127.6, 127.6, 127.9, 128.3, 128.4, 128.4,
128.5, 128.6, 128.7, 128.8, 128.9, 129.0, 130.2, 135.5, 135.8, 136.0,
136.0, 136.2, 146.9, 146.9, 149.1, 151.6, 151.8, 164.2, 164.9, 164.9,
165.9, 168.9, 169.1, 169.1. HRMS (ESI): [M+Na]+ m/z calcd,
1476.5323; found, 1476.5345.
Vancomycin-PEG-Ent (42). A DMSO solution of 41 (19 mg/mL,
1.3 mM, 250 μL), an aqueous solution of 8 (20 mg/mL, 1.3 mM, 250
μL), a DMF solution of benzoic acid (49 mg/mL, 450 mM, 50 μL),
and an aqueous solution of CuSO4 (10 mg/mL, 45 mM, 50 μL) were
mixed together, and an additional 400 μL of DMSO was added to
yield a clear light blue solution. An aqueous solution of sodium
ascorbate (NaAsc, 18 mg/mL, 90 mM, 50 μL) was subsequently
added. The reaction turned from colorless to yellow and was stirred at
rt for 15 min, at which time another 50 μL of aqueous NaAsc was
added. After stirring for 15 min, the crude reaction was frozen and
lyophilized to give a brown oil. The oil was dissolved in a 2:1:1 ratio of
dioxane/MeOH/H2O and purified by semipreparative HPLC (50% B
for 5 min followed by 50−100% B over 11 min, 4 mL/min). The
benzyl-protected precursor of 42 eluted at 13 min and was obtained as
white solid after lyophilization (3.5 mg, 36%). HRMS (ESI): [M
+2Na]2+/2 m/z calcd, 1520.5030; found, 1520.5171.
A portion of this precursor (14 mg, 4.7 μmol; obtained from four
250-μL scale click reactions) was dissolved in 30% H2O/MeOH, the
flask was purged with N2, and 16 mg of Pd/C (10 wt %) was added.
The reaction was stirred under H2 (1 atm) for 24 h at rt, and the Pd/C
was removed by centrifugation (13,000 rpm, 10 min). The supernatant
was concentrated by lyophilization and the resulting residue was
dissolved in a 2:1:1 mixture of dioxane/MeOH/H2O. HPLC
purification (20% B for 5 min followed by 20−46% B over 8 min, 4
mL/min) gave 43 as white solid (6.3 mg, 55%). The HPLC trace of
the purified product is reported in Supporting Information. HRMS
(ESI): [M+2H]2+/2 m/z calcd, 1228.3796; found, 1228.3796.
RESULTS AND DISCUSSION
■
Design and Synthesis of Monofunctionalized Entero-
bactin Platforms. We present a stepwise synthesis to
monofunctionalized enterobactin scaffolds in Scheme 1. Guided
by the structures of MGE and DGE (Figure 1A), we chose to
install functional groups amenable to synthetic modification at
the C5 position of one enterobactin catechol ring. This position
is remote from the iron-binding hydroxyl groups in addition to
the macrolactone (Figure 2). Prior studies of the salmochelins
indicate that modification at this site compromises neither
18393
dx.doi.org/10.1021/ja3077268 | J. Am. Chem. Soc. 2012, 134, 18388−18400

Journal of the American Chemical Society
Article
a
Scheme 2. Syntheses of Enterobactin−Cargo Conjugates 29−35
a
The syntheses of 14−18 are provided as Supporting Information.
Fe(III) complexation nor the esterase-catalyzed hydrolysis of
the macrolactone.33,64 The structure of the antibiotic−side-
rophore conjugate MccE492m (Figure S1),65 which exhibits a
monoglucosylated enterobactin derivative attached to a
ribosomal peptide, also influenced our decision to prepare
monofunctionalized enterobactin platforms. We selected
methyl 5-allyl-3-methoxysalicylate 19 as a starting material
because of its commercial availability. This precursor was
demethylated using BBr3 in the presence of DIPEA to prevent
HBr addition to the alkene moiety, and 20 was obtained in 53%
yield as a white powder. Benzyl protection of 20 and
subsequent hydrolysis of the methyl ester in refluxing sodium
hydroxide were performed following a literature protocol62 for
catecholate protection of DHB and 21 was obtained in 99%
yield as a white powder. Palladium-catalyzed isomerization of
the alkene was achieved by using PdCl2 in degassed methanol
and 22 was obtained in 89% yield as a light yellow solid
following workup. Next, a one-pot coupling reaction between
the enterobactin trilactone 4, 6, and 22 was performed with a
1:1.5:1 ratio and PyAOP as the coupling reagent. This reaction
provided a mixture of 23, its di- and trisubstituted analogues,
and unmodified Ent. These products were separated by flash
chromatography and afforded 23 in 37% yield as a white foam.
The 1:1.5:1 ratio of 4/6/22 was chosen based on several
optimization trials and this ratio provided the highest yield of
the desired monosubstituted product. Oxidation of alkene 23
by using OsO4 and NaIO4 in mixed 1,4-dioxane/water afforded
24 as a white foam in 58% yield. Further oxidation of 24 under
mild conditions provided carboxylic acid 25 in 76% yield as a
white powder. This stepwise synthesis provides gram quantities
of 23−25 (L-isomers) in high purity, and these molecules are
stable when stored as dry solids at 4 °C. The stepwise coupling
and oxidations were also performed using triserine lactone 5 to
afford the D-enantiomers alkene 26, aldehyde 27, and acid 28
(Supporting Information). The D-enantiomer of Ent is
transported into E. coli by FepA, but it is not a substrate for
the enterbactin esterase Fes.66 We therefore reasoned that
conjugates based on D-Ent would provide useful controls for
conjugate uptake studies, and that this enantiomer may also be
advantageous in antibacterial drug delivery applications because
it provides an iron-starvation effect.
This synthesis provides a family of enterobactin scaffolds
amenable to functionalization. For instance, alkene 23 may be
employed in olefin cross metathesis,67 aldehyde 24 in reductive
amination, and acid 25 in peptide coupling reactions.
Moreover, other organic transformations of 22 or 23 may
provide additional versatile functional groups (e.g., hydroxyl),
affording more synthetic possibilities for enterobactin deriva-
tives that can be utilized in various applications.
Design and Synthesis of Enterobactin−Cargo Con-
jugates. We selected carboxylic acid 25 as a key intermediate
for the preparation of enterobactin−cargo conjugates, and
evaluated two strategies for appending cargo to the entero-
bactin scaffold. In one thrust, standard peptide coupling
chemistry was employed to link cargo to the enterobactin
acid via an amide bond (Schemes 2 and 3). In the second
approach, enterobactin-azide 41 was prepared and “click”
chemistry utilized for cargo attachment (Scheme 4). In both
cases, benzyl deprotection unmasked the enterobactin
catecholates in the final step.
We selected a variety of commercially available molecules
housing carboxylic acids as cargos. The cargos presented in
Scheme 2 vary in size and shape and include a simple Boc
protecting group, cyclohexane, naphthalene, phenylmethyl-
benzene, ciprofloxacin, and coumarin 343. This selection
includes cargo expected to be nontoxic (e.g., Boc, cyclohexane)
in addition to an antibiotic (ciprofloxacin) and a fluorophore
(coumarin 343). Next, we selected PEG3 as a stable and water-
compatible linker. It provides ca. 14-Å separation between
enterobactin and the cargo. The conjugates depicted in Scheme
18394
dx.doi.org/10.1021/ja3077268 | J. Am. Chem. Soc. 2012, 134, 18388−18400

Journal of the American Chemical Society
Article
Scheme 3. Synthesis of Enterobactin−Ciprofloxacin Conjugate 40
Scheme 4. Syntheses of Enterobactin−Cargo Conjugates by Click Chemistry
2 were prepared by coupling the PEG-derivatized cargo 10,
14−18 to 25 using PyAOP as the coupling reagent. The
resulting benzyl-protected conjugates were purified by pre-
parative TLC and obtained in yields ranging from 26% (Bn-34)
to 75% (Bn-29). Benzyl deprotection reactions were performed
by hydrogenation over Pd/C and the resulting enterobactin−
cargo conjugates were purified by reverse-phase semiprepar-
ative HPLC. Conjugates 29−35 were obtained in milligram
quantities and high purity judging by analytical HPLC (Figures
S2−S11) and LC/MS analysis (Table S1). Conjugate 31
houses D-Ent and was prepared to probe the role of Fes-
mediated hydrolysis in the bacterial growth recovery assays.
Scheme 3 exhibits the synthesis of enterobactin−ciproflox-
acin 40 where the PEG linker is substituted by a C5 alkyl chain
to probe the consequences of variable linker composition and
hydrophilicity. The synthesis of 40 was carried out by reacting
ciprofloxacin with 6-Boc-aminohexanoic acid 12 followed by
Boc deprotection, coupling of the resulting free amine to 25,
and benzyl deprotection. The carboxylic acid of ciprofloxacin
was protected in situ by using trimethylsilyl chloride (TMSCl)
18395
dx.doi.org/10.1021/ja3077268 | J. Am. Chem. Soc. 2012, 134, 18388−18400

Journal of the American Chemical Society
Article
Figure 3. E. coli ATCC 33475 (ent-) and P. aeruginosa PAO1 (pvd-, pch-) growth recovery assays employing L-Ent and select enterobactin−cargo
conjugates (50% MHB, 200 or 600 μM DP, t = 19 h, 30 °C). Gray bars: OD600 of bacteria cultured in the absence of DP. Black bars: OD600 of
bacteria cultured in the presence of 200 (E. coli) or 600 (P. aeruginosa) μM DP. (A) L-Ent promotes growth recovery of E. coli. (B) Enterobactin
conjugate 30 housing a cyclohexyl cargo affords growth recovery of E. coli. (C) Enterobactin conjugate 34 housing a coumarin moiety affords little-
to-no growth recovery of E. coli. (D) L-Ent promotes growth recovery of P. aeruginosa. (E) Enterobactin conjugate 30 housing a cyclohexyl cargo
affords growth recovery of P. aeruginosa. (F) Enterobactin conjugate 34 housing a coumarin moiety affords growth recovery of P. aeruginosa. Figures
4 and S15 contain the assay results for the other conjugates. Each bar indicates the average of three independent replicates (two wells per replicate)
and the error bars are the standard deviation of the mean.
to prevent self-coupling in the syntheses of both 35 and 40. In
this general approach of attaching a carboxylic acid cargo, the
linkers were first coupled to the cargo rather than to the Ent
scaffold because the Ent macrolactone degrades in the presence
of primary amines and under acidic conditions such as those
required to remove Boc protecting groups.
conjugate exhibited catecholate absorption at ca. 316 nm
(MeOH, rt). With the exception of 34, which afforded a yellow
solution because of the coumarin moiety, methanolic solutions
of each conjugate turned from colorless to wine-colored
following the addition of ca. one equiv of aqueous Fe(III), and
the expected ligand-to-metal charge transfer (LMCT) bands
were observed, indicating Fe(III) coordination to the entero-
bactin catecholates (Figures S12−S14).70
Enterobactin−Cargo Conjugate Delivery to the E. coli
Cytoplasm. We employed three nonpathogenic E. coli strains
that are defective in enterobactin synthesis, enterobactin
transport, or ferric enterobactin utilization in growth recovery
assays (Table S2). E. coli ATCC 33475 (ent-) cannot
biosynthesize enterobactin, but retains the capacity to import
and metabolize the siderophore.71 E. coli H1187 (fepA-) lacks
the outer membrane enterobactin receptor. E. coli K-12
JW0576 (fes-) can accumulate ferric enterobactin, but cannot
release the iron because it is deficient in the enterobactin
esterase Fes. As a result of these defects in iron metabolism, all
three strains grow poorly under conditions of iron limitation.71
The iron chelator DP was used to generate iron-deficient
conditions and promote expression of siderophore transport
machinery in the growth recovery assays.
In Scheme 4, we present the synthesis of 42, an
enterobactin−vancomycin conjugate assembled via click
chemistry. Vancomycin is a nonribosomal peptide antibiotic
active against Gram-positive organisms that inhibits cell wall
biosynthesis by binding to the D-Ala-D-Ala of lipid II and
blocking peptidoglycan cross-linking.68 It exhibits poor activity
against Gram-negative bacteria because it is too large to cross
the outer membrane. Because modification of the C-terminal
carboxylic acid with a PEG chain did not perturb its
antibacterial activity,69 we selected this site as a point of
attachment. Moreover, we employed click chemistry for the
conjugate assembly to avoid complications with the various
functional groups exhibited by vancomycin. First, the azide-
functionalized PEG linker 11 was coupled to 25 to afford the
enterobactin-azide 41 in 68% yield. Copper(I)-catalyzed azide−
alkyne cycloaddition of 41 with alkyne 863 subsequently
afforded enterobactin−vancomycin 42 in 55% yield following
hydrogenation and purification. This synthetic approach was
extended to 43, a small analogue of 42 that houses a tert-butyl
cargo, and the strategy is also applicable to other alkyne-
substituted cargos that are compatible with the benzyl
deprotection conditions.
We first evaluated whether the enterobactin conjugates
afforded growth recovery of E. coli (ent-) cultured under iron-
deficient conditions (50% MHB, 200 μM DP). E. coli (ent-)
grew to OD600 ≈ 0.35 in 50% MHB medium (30 °C, t = 19 h),
and this value decreased to <0.05 when 200 μM DP was added
to the media. Low-micromolar concentrations of L-Ent restored
Enterobactin−Cargo Conjugates Coordinate Fe(III).
The optical absorption spectrum of each enterobactin−cargo
growth, as expected,70 and the E. coli cultures reached OD600
≈
18396
dx.doi.org/10.1021/ja3077268 | J. Am. Chem. Soc. 2012, 134, 18388−18400

Journal of the American Chemical Society
Article
Figure 5. Comparison of growth recovery for E. coli (ent-), E. coli
(fepA-), and E. coli (fes-) with conjugate 29 in the presence of 200 μM
DP. Black bars: E. coli (fes-) cultured with conjugate 29. White bars: E.
coli (fepA-) cultured with conjugate 29. Gray bars: E. coli (ent-)
cultured with conjugate 29 in the presence of 200 μM DP. See Figures
4 and S16 for additional data.
Figure 4. Comparative effects of enterobactin−cargo conjugates on
bacterial cell growth. E. coli and P. aeruginosa were cultured in the
presence of 10 μM of L-Ent 1, D-Ent 9 and the enterobactin−cargo
conjugates 29−35, 40, 42, 43 in the absence (gray bars) and presence
(black bars) of DP (50% MHB, T = 30 °C, t = 19 h). (A) E. coli
ATCC 33475 (ent-) and the DP concentration was 200 μM. (B) P.
aeruginosa PAO1 (pvd-, pch-) and the DP concentration was 600 μM.
NC refers to a no-conjugate control.
of active transport into E. coli, resulting in extracellular iron
chelation and hence nutrient deprivation. Taking all observa-
tions into account, including those for P. aeruginosa described
below, we contend that the latter option is the most probable
explanation.
Enterobactin−Cargo Conjugate Delivery to the P.
aeruginosa Cytoplasm. We next sought to determine
whether the enterobactin−cargo conjugates provided growth
recovery for P. aeruginosa PAO1. This Gram-negative
opportunistic human pathogen synthesizes and exports two
siderophores, pyoverdine (pvd) and pyochelin (pch), and
employs multiple additional mechanisms for iron acquisi-
tion.72,73 P. aeruginosa utilizes enterobactin as a xenosidero-
phore, and the genes pfeA74,75 and pirA76 encode outer
membrane enterobactin transporters. Similar to the E. coli
experiments, we focused on using P. aeruginosa strains deficient
in siderophore production or utilization in growth recovery
assays. P. aeruginosa K648 (pvd-, pch-) is deficient in both
pyoverdine and pyochelin biosynthesis and shows attenuated
growth in iron-deficient conditions, whereas P. aeruginosa K407
(pvd-, pFr-) is deficient in pyoverdine biosythesis and lacks the
enterobactin transporter PfeA.74
In 50% MHB medium, P. aeruginosa (pvd-, pch-) grew to
OD600 ≈ 0.45 (30 °C, t = 19 h) and this value diminished to ca.
0.25 in the presence of 600 μM DP. Supplementation of the
iron-limiting growth medium with low-micromolar concen-
trations of L-Ent resulted in the restoration of P. aeruginosa
growth to OD600 ≈ 0.40 (Figure 3). Comparable growth
recovery was observed for cultures treated with eight of the
nine conjugates based on L-Ent (Figures 3, 4, and S18).
Vancomycin 42, which exhibits the largest cargo, afforded a
growth inhibitory effect ( DP) as observed for E. coli (ent-). In
contrast to its L-Ent analogue 30, conjugate 31 based on D-Ent
was growth inhibitory as was D-Ent (Figures 4 and S18). This
result demonstrates that P. aeruginosa also requires the L-isomer
for iron utilization. Lastly, no growth enhancement of P.
aeruginosa (pvd-, pFr-) was observed in the presence of L-Ent or
conjugate 30 (600 μM DP); instead, these siderophores caused
growth inhibition at micromolar concentrations (Figure S19).
These results demonstrate that PfeA is necessary for conjugate-
mediated growth recovery, supporting its role as a transporter
for the enterobactin conjugates. In total, these assays reveal that
the enterobactin transport machinery of P. aeruginosa, and PfeA
0.2 in the presence of 10 μM Ent (Figure 3). Likewise, low-
micromolar concentrations of the enterobactin−cargo con-
jugates 29−33 and 43 exhibiting Boc (29, 43), cyclohexyl (30),
napthyl (32), and phenylmethylbenzyl (33) cargos afforded
growth recovery to similar levels (Figures 4 and S15). No
growth restoration was observed when E. coli (fepA-) or E. coli
(fes-) were cultured with 29 or 30 (Figures 5 and S16), which
supports the notion that the growth recovery of E. coli (ent-)
results from FepA-mediated cytoplasmic transport and Fes-
catalyzed hydrolysis of the enterobactin moiety to release iron.
Moreover, the D-enantiomer of enterobactin, D-Ent 9, is not a
substrate for Fes and does not provide growth recovery
(Figures 4 and S17).66 Indeed, no growth promotion occurred
when E. coli (ent-) was treated with conjugate 31, the D-
enantiomer of 30 (Figures 4 and S15). Taken together, these
results demonstrate that the enterobactin transport machinery
has the capacity to recognize and transport cargo-derivatized
enterobactin scaffolds to the E. coli cytoplasm, and that these
molecules are substrates for the cytoplasmic esterase Fes.
We observed no convincing evidence for marked uptake of
larger cargos by E. coli ATCC 33475, which suggests that FepA
of this E. coli strain has a cargo size limit. For instance, under
iron limitation, negligible E. coli growth recovery and no
toxicity was observed following treatment with the entero-
bactin−coumarin conjugate 34 (Figure 3), suggesting that E.
coli (ent-) may not readily import 34. Moreover, no growth
recovery occurred following treatment of E. coli with either
ciprofloxacin 35 or 40 (Figures 4 and S15). In the absence of
DP, these conjugates afforded a concentration-dependent
inhibition of E. coli growth. Likewise, 10 μM of vancomycin
42 inhibited the growth of E. coli ( DP, Figures 4 and S15).
This behavior contrasts that of unmodified vancomycin, which
is inactive against E. coli over the concentration range employed
in this study. Two possible origins for inhibitory activity of the
ciprofloxacin and vancomycin conjugates are (i) enterobactin−
antibiotic uptake and resulting antibacterial action or (ii) a lack
18397
dx.doi.org/10.1021/ja3077268 | J. Am. Chem. Soc. 2012, 134, 18388−18400

Journal of the American Chemical Society
Article
in particular, recognizes and delivers various cargo-modified
enterobactin scaffolds to the cytoplasm.
observations that catechol-modified β-lactams were recognized
by the iron-uptake machinery of Gram-negative microbes.43−46
Significant efforts have been made to prepare and characterize
synthetic siderophore-antibiotic conjugates with the goal of
targeting drug-resistant Gram-negative pathogens.13,14 Timely
examples of siderophore−antibiotic conjugates with antimicro-
bial activity include a mycobactin-artemisinin conjugate that
kills Mycobacterium tuberculosis and Plasmodium falciparum,82
and amoxicillin/ampicillin-appended tripodal triscatecholates
that exhibit potent antipseudomonal activity relative to the
parent β-lactam antibiotics.49 One bottleneck with this general
approach, and using siderophores in other applications, is that
few synthetically tractable and modifiable native siderophores
are available. DFO B and pyoverdine, which are readily
obtained commercially (DFO B) or from bacterial cultures
(pyoverdines), provide free amino groups useful for con-
jugation and are most commonly derivatized for application-
based work.18 Syntheses of modified pyochelin,15 petrobactin,19
and mycobactin82,83 platforms that house functional groups
amenable to site-specific elaboration have been reported, and
these scaffolds are important contributions to the toolkit of
siderophores that can be modified without compromising
Fe(III) coordination in addition to recognition by siderophore-
binding proteins. The syntheses described in this work provide
enterobactin with a functional handle for versatile chemical
modifications, and will allow strategic use of this canonical
siderophore in a multitude of chemical biology and
biotechnology initiatives.
Unanswered questions regarding the antibacterial activity and
fate of reported synthetic siderophore-antibiotic conjugates
exist. Whether a given conjugate is actively transported into the
bacterial cell is oftentimes unclear. Because FepA recognizes
relatively large biomolecules including MccE492m (84-aa) and
colicin B (324-aa), it is tempting to predict that FepA may
accommodate almost any cargo appended to an enterobactin or
catecholate platform. The results presented in this work
challenge this notion and indicate that cargo size is an
important and species-specific parameter. Our assays indicate
that P. aeruginosa PAO1 has a greater capacity to import
enterobactin−cargo conjugates than E. coli ATCC 33475. It will
be interesting to determine the cargo scope of other E. coli
strains and bacterial species that utilize enterobactin for iron
acquisition, and understand the molecular and physiological
basis for such variations. Colicins are largely α-helical40 and
MccE492m shares some sequence homology with colicins.84 It
is likely that some enterobactin receptors have decreased
propensity to transport synthetic small molecules or natural
products with less structural malleability (i.e., vancomycin) than
an α-helical peptide. Moreover, factors such as electrostatics
and functional groups may also influence transport.
Ciprofloxacin is a fluoroquinolone antibiotic that acts in the
cytoplasm and inhibits DNA gyrase.77 The fact that
ciprofloxacin conjugates 35 and 40 each restored P. aeruginosa
growth indicated that the cargo was successfully delivered to
the cytoplasm of this microbe with negligible impact of the
variable linker composition, and that conjugation of cipro-
floxacin to enterobactin attenuated its antibacterial activity.
This observation is in general agreement with reports of
pyoverdine−fluoroquinolone78 and pyochelin−fluoroquino-
lone79,80 conjugates where the antibiotic was covalently
attached to the siderophore and points to the need for
appropriate linker design for fluoroquinolone delivery and
release after cellular entry.81 These pyoverdine/pyochelin−
antibiotic conjugates afforded no antipseudomonal activity or
diminished activity relative to the unmodified drug, and the
pyoverdine−fluoroquinolone antibiotic exhibited decreased E.
coli gyrase inhibitory activity in vitro.78
A comparison of the enterobactin−cargo growth recovery
profiles for E. coli and P. aeruginosa (Figures 4, S15, and S18)
reveals that these particular microbes have different capacities
for internalizing enterobactin−cargo conjugates, and that cargo
size is an important factor. Vancomycin has a rigid dome-like
structure and a molecular weight of ca. 1.4 kDa, and the assays
presented in this work suggest that this molecule is too big for
enterobactin-mediated transport into E. coli or P. aeruginosa. In
contrast, small and malleable cargos such as a Boc protecting
group and cyclohexane afforded growth recovery comparable to
that of L-Ent for both strains. A comparison of OD600 values for
bacterial cultures treated with such conjugates (e.g., 29, 30, 32,
34) shows that growth recovery to levels comparable to that of
L-Ent occurs at a conjugate concentration of 1 μM for P.
aeruginosa whereas 10 μM is required for E. coli. P. aeruginosa
responds to lower Ent concentrations than E. coli, which
indicates a higher uptake efficiency. Coumarin 343 is an
example of a cargo that exhibits no signs of toxicity over the
concentration range tested and affords markedly different
results on microbial growth promotion for these two species. A
comparison of the ciprofloxacin conjugate data for E. coli and P.
aeruginosa also suggests differential uptake. For both the
ciprofloxacin and coumarin cargo, the growth recovery assays
indicate that the enterobactin transport machinery of P.
aeruginosa imports these cargos whereas the E. coli system
does not. These observations suggest that species-selective
targeting may be possible with strategic cargo choice even when
a siderophore is utilized by multiple microbial species.
SUMMARY AND PERSPECTIVES
■
We have designed and prepared a family of monofunctionalized
enterobactin derivatives, and utilized these scaffolds for the
preparation of enterobactin−cargo conjugates bearing cargos of
varying size and complexity. Growth recovery assays employing
E. coli and P. aeruginosa revealed that the enterobactin uptake
machineries of these Gram-negative species recognize and
transport enterobactin−cargo conjugates to the Gram-negative
cytoplasm. These studies are significant in several respects.
First, the notion of using siderophores for antibiotic delivery
across the Gram-negative outer membrane, which serves as a
permeability barrier, has achieved long-term interest.6−8,13 Such
“Trojan horse” antibiotics are largely inspired by the side-
romycins,11,12 a family of siderophore-antibiotic conjugates
produced by the soil bacterium Streptomyces, and by early
The mechanisms of iron release from siderophores, which
vary tremendously for the myriad of siderophores produced by
different bacterial species, are another important consideration
in siderophore−cargo conjugate design. Guided by studies of
chiral recognition in enterobactin transport, which demon-
strated that D-Ent is transported into E. coli but cannot be
hydrolyzed by Fes,71 we designed the monofunctionalized D-
Ent scaffolds to probe cytosolic delivery. This design feature
prevents esterase-catalyzed iron release from enterobactin-
based conjugates in the cytoplasm and may have practical
utility. From the standpoint of drug delivery, a tug-of-war may
result from utilizing an iron-supplying siderophore that confers
a growth advantage for delivering a toxic payload to a bacterial
18398
dx.doi.org/10.1021/ja3077268 | J. Am. Chem. Soc. 2012, 134, 18388−18400

Journal of the American Chemical Society
Article
(8) Budzikiewicz, H. Curr. Top. Med. Chem. 2001, 1, 73−82.
cell, and preventing iron release may be beneficial. In other
applications, siderophore-fluorophore conjugates are of interest
for bacterial detection and diagnostics, and Fe(III) binding to
and release from the siderophore will likely influence the
photophysical properties of such molecules.
In summary, these investigations reveal that the enterobactin
transport machineries of E. coli (e.g., FepABCDG and TonB-
ExbB-ExbD) and P. aeruginosa (e.g., PfeA) will deliver
enterobactin-modified cargo to the Gram-negative cytoplasm.
Moreover, the preparative work affords a new siderophore
platform amenable to synthetic elaboration and an entry route
for employing the native enterobactin scaffold in a variety of
application-based initiatives that include intracellular cargo
delivery, iron sensing, siderophore labeling, protein and
pathogen detection, and therapeutic development.
(9) Ballouche, M.; Cornelis, P.; Baysse, C. Recent Pat. Anti-Infect.
Drug Discovery 2009, 4, 190−205.
(10) Mollmann, U.; Heinisch, L.; Bauernfeind, A.; Kohler, T.; Ankel-
̈
̈
Fuchs, D. Biometals 2009, 22, 615−624.
(11) Braun, V.; Pramanik, A.; Gwinner, T.; Koberle, M.; Bohn, E.
̈
Biometals 2009, 22, 3−13.
(12) Braun, V. Drug Resist. Update 1999, 2, 363−369.
(13) Ji, C.; Juar
2012, 4, 297−313.
́ ́
ez-Hernandez, R. E.; Miller, M. J. Future Med. Chem.
(14) Miller, M. J.; Zhu, H.; Xu, Y.; Wu, C.; Walz, A. J.; Vergne, A.;
Roosenberg, J. M.; Moraski, G.; Minnick, A. A.; McKee-Dolence, J.;
Hu, J.; Fennell, K.; Dolence, E. K.; Dong, L.; Franzblau, S.; Malouin,
F.; Mollmann, U. Biometals 2009, 22, 61−75.
̈
(15) Noel, S.; Guillon, L.; Schalk, I. J.; Mislin, G. L. A. Org. Lett.
̈
2011, 13, 844−847.
́
(16) Esposito, B. P.; Epsztejn, S.; Breuer, W.; Cabantchik, Z. I. Anal.
Biochem. 2002, 304, 1−18.
ASSOCIATED CONTENT
* Supporting Information
■
(17) Lam, C. K. S. C. C.; Jickells, T. D.; Richardson, D. J.; Russell, D.
A. Anal. Chem. 2006, 78, 5040−5045.
S
(18) Zheng, T.; Nolan, E. M. Metallomics 2012, 4, 866−880.
(19) Bugdahn, N.; Peuckert, F.; Albrecht, A. G.; Miethke, M.;
Syntheses and characterization of 14−18 and 26−28, general
liquid chromatography, mass spectrometry and microbiology
methods, summary of enterobactin−cargo conjugate character-
ization (Table S1), summary of bacterial strains and sources
(Table S2), structure of MccE492m (Figure S1), HPLC traces
for the purified conjugates (Figures S2−S11), optical
absorption spectra (Figures S12−S14), growth recovery assays
(Figures S15−19), 1H and 13C NMR spectra, and IR
spectroscopic data. This material is available free of charge
via the Internet at http://pubs.acs.org.
Marahiel, M. A.; Oberthur, M. Angew. Chem., Int. Ed. 2010, 49,
̈
10210−10213.
(20) Doorneweerd, D. D.; Henne, W. A.; Reifenberger, R. G.; Low,
P. S. Langmuir 2010, 26, 15424−15429.
(21) Kim, Y.; Lyvers, D. P.; Wei, A.; Reifenberger, R. G.; Low, P. S.
Lab Chip 2012, 12, 971−976.
(22) Raymond, K. N.; Dertz, E. A.; Kim, S. S. Proc. Natl. Acad. Sci. U.
S. A. 2003, 100, 3584−3588.
(23) Crosa, J. H.; Walsh, C. T. Microbiol. Mol. Biol. Rev. 2002, 66,
223−249.
AUTHOR INFORMATION
Corresponding Author
lnolan@mit.edu
(24) Loomis, L. D.; Raymond, K. N. Inorg. Chem. 1991, 30, 906−911.
(25) Buchanan, S. K.; Smith, B. S.; Venkatramani, L.; Xia, D.; Esser,
L.; Palnitkar, M.; Chakraborty, R.; van der Helm, D.; Deisenhofer, J.
Nat. Struct. Biol. 1999, 6, 56−63.
■
Notes
(26) Newton, S. M. C.; Igo, J. D.; Scott, D. C.; Klebba, P. E. Mol.
Microbiol. 1999, 32, 1153−1165.
The authors declare no competing financial interest.
(27) Stephens, D. L.; Choe, M. D.; Earhart, C. F. Microbiology-UK
1995, 141, 1647−1654.
ACKNOWLEDGMENTS
■
(28) Chenault, S. S.; Earhart, C. F. Mol. Microbiol. 1991, 5, 1405−
1413.
The Searle Scholars Program (Kinship Foundation), the
Department of Chemistry and the Undergraduate Research
Opportunities Program (UROP) at MIT, and the Amgen
Scholars Program (J.L.B.) are gratefully acknowledged for
financial support. We thank Prof. Keith Poole for providing the
P. aeruginosa strains employed in this work, Prof. Klaus Hantke
for providing the E. coli H1178 (fepA-) strain, Prof. Stephen J.
Lippard for use of an IR spectrophotometer and a melting
point apparatus, and Dr. Andrew Wommack and Mr. Phoom
Chairatana for carefully proof-reading the manuscript. E. coli K-
12 JW0576 was obtained from the Keio Collection.85 NMR
instrumentation maintained by the MIT DCIF is supported by
NSF grants CHE-9808061 and DBI-9729592.
(29) Shea, C. M.; McIntosh, M. A. Mol. Microbiol. 1991, 5, 1415−
1428.
(30) Chakraborty, R.; Storey, E.; van der Helm, D. Biometals 2007,
20, 263−274.
(31) Krewulak, K. D.; Vogel, H. J. Biochim. Biophys. Acta 2008, 1778,
1781−1804.
(32) Chu, B. C.; Garcia-Herrero, A.; Johanson, T. H.; Krewulak, K.
D.; Lau, C. K.; Peacock, R. S.; Slavinskaya, Z.; Vogel, H. J. Biometals
2010, 23, 601−611.
(33) Lin, H.; Fischbach, M. A.; Liu, D. R.; Walsh, C. T. J. Am. Chem.
Soc. 2005, 127, 11075−11084.
(34) Miethke, M.; Hou, J.; Marahiel, M. A. Biochemistry 2011, 50,
10951−10964.
(35) Baumler, A. J.; Norris, T. L.; Lasco, T.; Voigt, W.; Reissbrodt,
̈
REFERENCES
R.; Rabsch, W.; Heffron, F. J. Bacteriol. 1998, 180, 1446−1453.
(36) Lagos, R.; Baeza, M.; Corsini, G.; Hetz, C.; Strahsburger, E.;
Castillo, J. A.; Vergara, C.; Monasterio, O. Mol. Microbiol. 2001, 42,
229−243.
■
(1) Hider, R. C.; Kong, X. Nat. Prod. Rep. 2010, 27, 637−657.
(2) Miethke, M.; Marahiel, M. A. Microbiol. Mol. Biol. Rev. 2007, 71,
413−451.
(3) Rajkumar, M.; Ae, N.; Prasad, M. N. V.; Freitas, H. Trends
Biotechnol. 2010, 28, 142−149.
(37) Nolan, E. M.; Fischbach, M. A.; Koglin, A.; Walsh, C. T. J. Am.
Chem. Soc. 2007, 129, 14336−14347.
(38) Fischbach, M. A.; Lin, H.; Liu, D. R.; Walsh, C. T. Proc. Natl.
Acad. Sci. U. S. A. 2005, 102, 571−576.
(4) Bernhardt, P. V. Dalton Trans. 2007, 3214−3220.
(5) Manning, T.; Kean, G.; Thomas, J.; Thomas, K.; Corbitt, M.;
Gosnell, D.; Ware, R.; Fulp, S.; Jarrard, J.; Phillips, D. Curr. Med. Chem.
2009, 16, 2416−2429.
(6) Miller, M. J. Chem. Rev. 1989, 89, 1563−1579.
(7) Roosenberg, J. M., II; Lin, Y.-M.; Lu, Y.; Miller, M. J. Curr. Med.
Chem. 2000, 7, 159−197.
(39) Muller, S.; Valdebenito, M.; Hantke, K. Biometals 2009, 22,
̈
691−695.
(40) Cao, Z.; Klebba, P. E. Biochimie 2002, 84, 399−412.
́
(41) Lagos, R.; Tello, M.; Mercado, G.; Garcia, V.; Monasterio, O.
Curr. Pharm. Biotechnol. 2009, 10, 74−85.
18399
dx.doi.org/10.1021/ja3077268 | J. Am. Chem. Soc. 2012, 134, 18388−18400

Journal of the American Chemical Society
Article
(42) Rabsch, W.; Ma, L.; Wiley, G.; Najar, F. Z.; Kaserer, W.;
Schuerch, D. W.; Klebba, J. E.; Roe, B. A.; Gomez, J. A. L.; Schallmey,
M.; Newton, S. M. C.; Klebba, P. E. J. Bacteriol. 2007, 189, 5658−
5674.
(75) Dean, C. R.; Neshat, S.; Poole, K. J. Bacteriol. 1996, 178, 5361−
5369.
(76) Ghysels, B.; Ochsner, U.; Mollman, U.; Heinisch, L.; Vasil, M.;
̈
Cornelis, P.; Matthijs, S. FEMS Microbiol. Lett. 2005, 246, 167−174.
(77) Collin, F.; Karkare, S.; Maxwell, A. App. Microbiol. Biotechnol.
2011, 92, 479−497.
(43) Katsu, K.; Kitoh, K.; Inoue, M.; Mitsuhashi, S. Antimicrob. Agents
Chemother. 1982, 22, 181−185.
(44) Watanabe, N.-A.; Nagasu, T.; Katsu, K.; Kitoh, K. Antimicrob.
Agents Chemother. 1987, 31, 497−504.
(78) Hennard, C.; Truong, Q. C.; Desnottes, J.-F.; Paris, J.-M.;
Moreau, N. J.; Abdallah, M. A. J. Med. Chem. 2001, 44, 2139−2151.
́
(79) Rivault, F.; Liebert, C.; Burger, A.; Hoegy, F.; Abdallah, M. A.;
(45) Nakagawa, S.; Sanada, M.; Matsuda, K.; Hazumi, N.; Tanaka, N.
Antimicrob. Agents Chemother. 1987, 31, 1100−1105.
(46) Hashizume, T.; Sanada, M.; Nakagawa, S.; Tanaka, N. J. Antibiot.
1990, 43, 1617−1620.
Schalk, I. J.; Mislin, G. L. A. Bioorg. Med. Chem. Lett. 2007, 17, 640−
644.
(80) Noel, S.; Gasser, V.; Pesset, B.; Hoegy, F.; Rognan, D.; Schalk, I.
̈
J.; Mislin, G. L. A. Org. Biomol. Chem. 2011, 9, 8288−8300.
(81) Ji, C.; Miller, M. J. Bioorg. Med. Chem. 2012, 20, 3828−3836.
(82) Miller, M. J.; Walz, A. J.; Zhu, H.; Wu, C.; Moraski, G.;
(47) Mollmann, U.; Ghosh, A.; Dolence, E. K.; Dolence, J. A.; Ghosh,
̈
M.; Miller, M. J.; Reissbrodt, R. Biometals 1998, 11, 1−12.
(48) McKee, J. A.; Sharma, S. K.; Miller, M. J. Bioconjugate Chem.
1991, 2, 281−291.
Mollmann, U.; Tristani, E. M.; Crumbliss, A. L.; Ferdig, M. T.;
̈
Checkley, L.; Edwards, R. L.; Boshoff, H. I. J. Am. Chem. Soc. 2011,
133, 2076−2079.
(49) Ji, C.; Miller, P. A.; Miller, M. J. J. Am. Chem. Soc. 2012, 134,
9898−9901.
(83) Xu, Y.; Miller, M. J. J. Org. Chem. 1998, 63, 4314−4322.
(84) Pons, A.-M.; Zorn, N.; Vignon, D.; Delalande, F.; Van
Dorsselaer, A.; Cottenceau, G. Antimicrob. Agents Chemother. 2002,
46, 229−230.
(50) Diarra, M. S.; Lavoie, M. C.; Jacques, M.; Darwish, I.; Dolence,
E. K.; Dolence, J. A.; Ghosh, A.; Ghosh, M.; Miller, M. J.; Malouin, F.
Antimicrob. Agents Chemother. 1996, 40, 2610−2617.
(51) Ghosh, A.; Ghosh, M.; Niu, C.; Malouin, F.; Mollmann, U.;
̈
(85) Baba, T.; Ara, T.; Hasegawa, M.; Takai, Y.; Okumura, Y.; Baba,
M.; Datsenko, K. A.; Tomita, M.; Wanner, B. L.; Mori, H. Mol. Syst.
Biol. 2006, 2, No. 2006.0008.
Miller, M. J. Chem. Biol. 1996, 3, 1011−1019.
(52) Corey, E. J.; Bhattacharyya, S. Tetrahedron Lett. 1977, 45, 3919−
3922.
(53) Rastetter, W. H.; Erickson, T. J.; Venuti, M. C. J. Org. Chem.
1981, 46, 3579−3590.
(54) Shanzer, A.; Libman, J. J. Chem. Soc., Chem. Commun. 1983, 15,
846−847.
(55) Ramirez, R. J. A.; Karamanukyan, L.; Ortiz, S.; Gutierrez, C. G.
Tetrahedron Lett. 1997, 38, 749−752.
(56) Marinez, E. R.; Salmassian, E. K.; Lau, T. T.; Gutierrez, C. G. J.
Org. Chem. 1996, 61, 3548−3550.
(57) Rodgers, S. J.; Lee, C.-W.; Ng, C. Y.; Raymond, K. N. Inorg.
Chem. 1987, 26, 1622−1625.
(58) Tor, Y.; Libman, J.; Shanzer, A.; Felder, C. E.; Lifson, S. J. Am.
Chem. Soc. 1992, 114, 6661−6671.
(59) Ecker, D. J.; Loomis, L. D.; Cass, M. E.; Raymond, K. N. J. Am.
Chem. Soc. 1988, 110, 2457−2464.
(60) Stack, T. D. P.; Hou, Z.; Raymond, K. N. J. Am. Chem. Soc.
1993, 115, 6466−6467.
(61) Yu, X.; Dai, Y.; Yang, T.; Gagne,
2011, 67, 144−151.
́
M. R.; Gong, H. Tetrahedron
(62) Gardner, R. A.; Kinkade, R.; Wang, C.; Phanstiel, O., IV J. Org.
Chem. 2004, 69, 3530−3537.
(63) Arnusch, C. J.; Bonvin, A. M. J. J.; Verel, A. M.; Jansen, W. T.
M.; Liskamp, R. M. J.; de Kruijff, B.; Pieters, R. J.; Breukink, E.
Biochemistry 2008, 47, 12661−12663.
(64) Luo, M.; Lin, H.; Fischbach, M. A.; Liu, D. R.; Walsh, C. T.;
Groves, J. T. ACS Chem. Biol. 2006, 1, 29−32.
́
(65) Thomas, X.; Destoumieux-Garzon, D.; Peduzzi, J.; Afonso, C.;
Blond, A.; Birlirakis, N.; Goulard, C.; Dubost, L.; Thai, R.; Tabet, J.-C.;
Rebuffat, S. J. Biol. Chem. 2004, 279, 28233−28242.
(66) Abergel, R. J.; Zawadzka, A. M.; Hoette, T. M.; Raymond, K. N.
J. Am. Chem. Soc. 2009, 131, 12682−12692.
(67) Zheng, T.; Nolan, E. M. Unpublished results.
(68) Hubbard, B. K.; Walsh, C. T. Angew. Chem., Int. Ed. 2003, 42,
730−765.
(69) Lawson, M. C.; Shoemaker, R.; Hoth, K. B.; Bowman, C. N.;
Anseth, K. S. Biomacromolecules 2009, 10, 2221−2234.
(70) Scarrow, R. C.; Ecker, D. J.; Ng, C.; Liu, S.; Raymond, K. N.
Inorg. Chem. 1991, 30, 900−906.
(71) Wayne, R.; Frick, K.; Neilands, J. B. J. Bacteriol. 1976, 126, 7−
12.
(72) Mossialos, D.; Amoutzias, G. D. Future Microbiol. 2007, 2, 387−
395.
(73) Cornelis, P. App. Microbiol. Biotechnol. 2010, 86, 1637−1645.
(74) Poole, K.; Young, L.; Neshat, S. J. Bacteriol. 1990, 172, 6991−
6996.
18400
dx.doi.org/10.1021/ja3077268 | J. Am. Chem. Soc. 2012, 134, 18388−18400