Total synthesis and structure revision of mirubactin, and its iron binding activity

Article abstract of DOI:10.1246/cl.150520

Dechloro-chlorocatechelin A (3), a deschloro-derivative of a microbial siderophore chlorocatechelin A (2), was synthesized from 2,3-dihydroxybenzoic acid, D-arginine, and 1-benzyl Dglutamate. The spectral data were unambiguously identical with those of mi

Full text of DOI:10.1246/cl.150520

Received: May 28, 2015 | Accepted: June 20, 2015 | Web Released: June 27, 2015
CL-150520
Total Synthesis and Structure Revision of Mirubactin, and Its Iron Binding Activity
Shinji Kishimoto, Shinichi Nishimura, and Hideaki Kakeya*
Department of System Chemotherapy and Molecular Sciences, Division of Bioinformatics and Chemical Genomics,
Graduate School of Pharmaceutical Sciences, Kyoto University, Sakyo-ku, Kyoto 606-8501
(E-mail: scseigyo-hisyo@pharm.kyoto-u.ac.jp)
Dechloro-chlorocatechelin A (3), a deschloro-derivative of a
microbial siderophore chlorocatechelin A (2), was synthesized
from 2,3-dihydroxybenzoic acid, D-arginine, and 1-benzyl D-
glutamate. The spectral data were unambiguously identical with
those of mirubactin (1), revising its chemical structure. Mirubactin
showed potent iron binding affinity, which was comparable to that
of chlorocatechelin A (2).
difference between 1 and 2 is the location of the catecholate group:
2 has an acylguanidine structure composed of CDB and D-Arg,
not an O-acyl hydroxamic acid ester. However, the NMR and MS/
MS spectra of 1 closely resembled those of 2. These observations
led us to reinvestigate the chemical structure of mirubactin (1).
Here we report total synthesis of dechloro-chlorocatechelin A (3)
and structure revision of mirubactin (1) (Figure 1).
We planned to synthesize 3 in a similar way to 2.9 Scheme 1
shows the retrosynthesis of 3, which was fragmented into two
segments 4 and 5. The left segment 4 would be obtained by
coupling benzylated DHB 6 and D-Arg 7, while the right segment
5 can be synthesized from 1-benzyl D-glutamate 8 as reported
previously.9
Synthesis of the left segment 4 commenced from benzylation
of DHB 9. Compound 9 was reacted with BnBr, and then
hydrolyzed in alkaline conditions to give a carboxylic acid 6.
Protected DHB 6 was converted to acid chloride with (COCl)2
and subsequently reacted with D-Arg under Schotten-Baumann
conditions to yield the left segment 4 (34%) and the recovered
material 6 (50%) (Scheme 2).
Siderophores are low-molecular-weight metabolites that
microbes and plants excrete to acquire iron from the environ-
ment.1,2 Siderophores usually contain three bidentate ligands in a
molecule to form a stable octahedral Fe(III)-siderophore complex.
Bidentate groups in siderophores can be classified into three
types:3 catecholate type (e.g. 2,3-dihydroxybenzoic acid (DHB)
unit in enterobactin4), hydroxamate type (e.g. N-δ-hydroxy-N-δ-
formyl ornithine (hfOrn) unit in amychelin5), and α-hydroxy-
carboxylate type (e.g. citrate unit in rhizobactin 10216). Some
siderophores contain a single species of the bidentate group, while
others have two or three species.
In 2012, Marahiel and co-workers reported the isolation,
structure characterization, and biosynthesis of a siderophore
named mirubactin (1), from Actinosynnema mirum.7 This metab-
olite consists of two units of DHB, one unit each of D-arginine
(D-Arg) and D-hfOrn (Figure 1). The most characteristic feature
of this compound was that one unit of DHB was linked to
D-hfOrn by forming an unusual O-acyl hydroxamic acid ester.
However, this linkage was not determined with certainty by
chemical methods. Additionally, the O-acyl hydroxamic acid
ester disables the chelation ability of hydroxamate, which should
significantly decrease the affinity of the metabolite to Fe(III).
Recently, we reported the discovery and total synthesis of
chlorocatechelin A (2), a novel siderophore from Streptomyces sp.
ML93-86F2 (Figure 1).8,9 The substructures of 2 are similar to
those of 1, except for the fact that 2 contains 4-chloro-2,3-dihy-
droxybenzoic acid (CDB) instead of DHB. The major structural
The protected right segment 10 was prepared from 1-benzyl
D-glutamate in 6 steps (totally 70% yield) as described.9 The right
segment 5, which was obtained by removal of the Boc group of
3
H2N
NH
NH
R
N
NH2
NH
HO
OH
O
O
OH
O
O
4
5
H
N
HO
N
H
OH
OH
O
H
N
O
HO
R
O
N
OH
H
O
N
O
H
O
HO
N
H
OH
O
OH
2: R = Cl
3: R = H
1
6
7
8
Figure 1. Structures of mirubactin (1; originally reported), chloro-
catechelin A (2) and dechloro-chlorocatechelin A (3).
Scheme 1. Retrosynthesis of dechloro-chlorocatechelin A (3).
© 2015 The Chemical Society of Japan | 1303
Chem. Lett. 2015, 44, 1303–1305 | doi:10.1246/cl.150520

1. BnBr, K2CO3
DMF, r.t., 16.5 h
buffer (pH 7.0, I = 0.1 M, Figures 2a and 2b). An increasing
absorbance at 516 nm was observed until 1 equiv of FeCl3 was
added, indicating that 3 formed a complex with Fe(III) at a 1:1
ratio. Notably, the UV spectra of the synthesized compound 3 in
the absence and presence of Fe(III) resembled those reported for
natural mirubactin.7 Next, the redox potential of the Fe(III)-3
complex was measured in cyclic voltammetry (CV) experi-
ments.10 The mixture of 3 (200 ¯M) and FeCl3 (140 ¯M) in
50 mM BisTris buffer (pH 7.0, I = 0.1 M, Figure 2c) gave a
reversible voltammogram with a half wave potential (E1/2) of
¹550 mV (vs. NHE), which was slightly higher than that of
chlorocatechelin A (2, E1/2 = ¹578 mV)8 indicating that the iron
binding affinity of 3 is slightly lower than that of 2.10 These
results suggest that chloro substituents partly contribute to the
potency of the iron binding affinity of these metabolites.
OH
O
OBn O
HO
BnO
OH
OH
2. NaOH (aq)
MeOH, 90 °C, 72 h
96% (2 steps)
9
6
N
NH2
NH
BnO
1. (COCl)2, DMF (cat.)
DCM, r.t., 50 min
OBn O
2. D-Arg, NaOH (aq)
1,4-dioxane, r.t., 2 h
34% (2 steps)
OBn O
BnO
OH
N
H
50% recovery of 6
O
4
Scheme 2. Synthesis of the left segment 4.
In summary, we synthesized dechloro-chlorocatechelin A (3)
in 10 steps from 1-benzyl D-glutamate and revised the chemical
structure of mirubactin. Mirubactin (3) showed potent iron
binding properties, comparable to that of chlorocatechelin A (2).
10 with TFA, was condensed with the left segment 4 using HATU
and HOAt to give compound 11. Compound 11 was subjected to
hydrolysis followed by hydrogenolysis for removal of the benzyl
groups, yielding dechloro-chlorocatechelin A (3) (Scheme 3).
The NMR data of synthesized 3 were unambiguously identical
with those reported for mirubactin (Supporting Information,
Table S1),7 revealing that the chemical structure of mirubactin
should be 3.
We thank Dr. Kenji Kano (Kyoto University) for supporting
cyclic voltammetry experiments. This work was supported in part
by research grants from the Japan Society for the Promotion of
Science (JSPS), the Ministry of Health, Labour and Welfare of
Japan (MHLW), and the Ministry of Education, Culture, Sports,
Science and Technology of Japan (MEXT).
We finally investigated the iron binding properties of
synthesized mirubactin (3). We examined a spectrophotometric
titration experiment of compound 3 with FeCl3 in 50 mM BisTris
Supporting Information is available electronically on J-STAGE.
N
O
NH2
NH
N
O
NH2
NH
BnO
HO
O
H
O
O
OBn
OH
O
1. TFA, DCM
r.t., 30 min
N
OBn
Pd/C, H2
OBn O
O
OH
O
H
N
H
N
BnO
HO
THF, r.t. ,9.5 h
58%
2. 4, DIEA, HATU, HOAt
DMF, 0 °C to r.t., 3 h
N
H
OR
N
OH
OBn
H
BocHN
O
O
O
N
H
N
H
OBn
OH
10
3
11
12
R = Bn
R = H
LiOH (aq)
THF, r.t., 30 min
47% (3 steps)
Scheme 3. Synthesis of dechloro-chlorocatechelin A (3).
(a)
(b) (c)
0.4
0.3
0.2
0.1
0
-12
-8
-4
0
4
0
0.4
0.8
1.2
0
-0.4
-0.8
-1.2
Wavelength/nm
FeCl3/equiv
Potential/mV (vs. Ag|AgCl)
Figure 2. Iron binding properties of the synthesized compound 3. (a, b) Spectrophotometric titration of 3 with FeCl3 in 50 mM BisTris buffer
(pH 7.0, I = 0.1 M with NaCl). UVabsorption spectra are shown in (a); 0, 0.4, 0.6, 0.8, 1.0, and 1.1 equiv of FeCl3 were added (black to green). The
absorptions at 516 nm are plotted in (b). (c) Cyclic voltammogram of Fe(III)-3 complex (100 ¯M of 3 and 60 ¯M of FeCl3) in 50 mM BisTris buffer
(pH 7.0, I = 0.1 M with NaCl).
1304 | Chem. Lett. 2015, 44, 1303–1305 | doi:10.1246/cl.150520
© 2015 The Chemical Society of Japan

References
Jr., J. B. Neilands, J. Am. Chem. Soc. 1993, 115, 3950.
T. W. Giessen, K. B. Franke, T. A. Knappe, F. I. Kraas, M.
Bosello, X. Xie, U. Linne, M. A. Marahiel, J. Nat. Prod.
2012, 75, 905.
S. Kishimoto, S. Nishimura, A. Hattori, M. Tsujimoto, M.
Hatano, M. Igarashi, H. Kakeya, Org. Lett. 2014, 16, 6108.
S. Kishimoto, S. Nishimura, M. Hatano, M. Igarashi, H.
Kakeya, J. Org. Chem. 2015, 80, 6076.
1
S. C. Andrews, A. K. Robinson, F. Rodríguez-Quiñones,
7
FEMS Microbiol. Rev. 2003, 27, 215.
2
3
4
M. Sandy, A. Butler, Chem. Rev. 2009, 109, 4580.
R. C. Hider, X. Kong, Nat. Prod. Rep. 2010, 27, 637.
J. R. Pollack, J. B. Neilands, Biochem. Biophys. Res.
Commun. 1970, 38, 989.
8
9
5
6
M. R. Seyedsayamdost, M. F. Traxler, S.-L. Zheng, R. Kolter,
J. Clardy, J. Am. Chem. Soc. 2011, 133, 11434.
M. Persmark, P. Pittman, J. S. Buyer, B. Schwyn, P. R. Gill,
10 S. W. Taylor, G. W. Luther, III, J. H. Waite, Inorg. Chem.
1994, 33, 5819.
Chem. Lett. 2015, 44, 1303–1305 | doi:10.1246/cl.150520
© 2015 The Chemical Society of Japan | 1305