Synthesis of a novel azobenzene-based trihydroxamate siderophore and photoregulation of its ferric complex structure

Article abstract of DOI:10.1016/S0040-4039(00)00175-1

In order to develop an artificial photoresponsive siderophore by utilizing the photo-isomerization of an azobenzene group, a novel trihydroxamate siderophore based on an azobenzene skeleton was synthesized. The 1:1 and 2:2 stoichiometry of the ferric complex can be almost perfectly regulated reversibly by irradiation with UV and visible light, respectively. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)00175-1

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 2411–2414
Synthesis of a novel azobenzene-based trihydroxamate
siderophore and photoregulation of its ferric complex structure
Takeshi Masuda, Sayaka Asari, Seizo Tamagaki and Takeshi Nagasaki ∗
Department of Bioapplied Chemistry, Faculty of Engineering, Osaka City University, Sugimoto 3-3-138, Sumiyoshi-ku,
Osaka 5588585, Japan
Received 24 December 1999; revised 17 January 2000; accepted 21 January 2000
Abstract
In order to develop an artificial photoresponsive siderophore by utilizing the photo-isomerization of an azoben-
zene group, a novel trihydroxamate siderophore based on an azobenzene skeleton was synthesized. The 1:1 and 2:2
stoichiometry of the ferric complex can be almost perfectly regulated reversibly by irradiation with UV and visible
light, respectively. © 2000 Elsevier Science Ltd. All rights reserved.
Keywords: diazo compound; hydroxamic acid; photochemistry; siderophore.
Photoresponsive systems play important roles in biological process, such as photosynthesis,1 vision,2
phototropism,3 and phototaxis.4 One common attribute of every system is the regulation of activity
of a biologically active macromolecule by a low molecular weight photochromic molecule capable
of assuming at least two states, whose relative concentrations are determined by light of specific
wavelengths. To apply this concept of naturally occurring processes to chemical processes, a great deal of
effort has been made on the development of artificial photoregulation systems such as ionophores,5 sugar-
receptors,6 macrocyclic host-molecules,7 oligonucleotides,8 enzyme-inhibitors,9 enzymes themselves,10
and catalysts for chemical reactions.11 On the other hand, a variety of microorganisms produce low
molecular weight iron carrier molecules called siderophores, which solubilize iron and make it available
to iron-deficient cells. Herein we report the synthesis and properties of a novel and photoresponsive
artificial siderophore based on an azobenzene skeleton which changes the structure of the ferric complex
in response to photo-irradiation.
The synthetic procedure for Azo-Gly is depicted in Scheme 1. Since azobenzene is easily reduced to
the hydrazine under the reductive conditions for deprotection of a benzyl group (a useful protecting group
for a hydroxamic acid), direct acylation of an amino group involved in a hydroxamate derivative was
chosen for linking of hydroxamate units to an azobenzene skeleton. The photoresponsive siderophore,
Azo-Gly, was finally prepared in moderate yield (52%) by a coupling reaction between azobenzene-
based tricarboxylic acid (IV) and glycine-N-hydroxyl-N-methylamide in the presence of DCC and HOBt
∗
Corresponding author. E-mail: nagasaki@bioa.eng.osaka-cu.ac.jp (T. Nagasaki)
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)00175-1
tetl 16455

2412
followed by HPLC purification (ODS column; eluents: H2O/CH3CN=50/50). The product was identified
1
by IR, H NMR, and high resolution-positive FAB-mass spectroscopies and elemental analysis.† The
absorption spectrum of trans-Azo-Gly in DMF had an absorption peak at 354 nm with a molar extinction
coefficient of 31 000. Exposure to UV light by a high-pressure Hg-lamp (400 W) through a visible
filter (300<λ<400 nm) converted 86% of trans-isomer to cis-Azo-Gly. This photostationary state was
achieved in 90 s. The cis-isomer exhibits an absorption maxima at 442 nm (ε=2 440). For the experiment
described below, we used the cis/trans (86/14) mixture as cis-Azo-Gly induced by UV irradiation. When
the mixture was further irradiated with visible light (λ>400 nm), the cis-isomer isomerized back to the
trans form instantly. In contrast, cis–trans thermal isomerization of Azo-Gly was very slow (half-life 20
h) at 30°C. Thus, the decrease in the content of the cis-isomer by the thermal isomerization was negligible
during the measurement of spectroscopies of cis-Azo-Gly.
Scheme 1. (i) SOCl2, diethyl iminodiacetate, Et3N (91%); (ii) Cs2CO3, ethyl bromoacetate (87%); (iii) H2O, OH− then H+
(96%); (iv) DCC, HOBt, glycine-N-hydroxyl-N-methylamide hydrochloride, Et3N (52%)
As the solubility of Azo-Gly in water was too low, its FeIII-chelating properties were investigated
in DMF in the presence of triethylamine. In the UV–vis spectrum of a 1:1 molar mixture of FeIII and
cis-Azo-Gly in the presence of fourfold excess of triethylamine, the characteristic LMCT (ligand to
metal charge transfer) band was observed at 420 nm (ε=2 750). The λmax and ε value suggested the
intramolecular formation of the 1:3 complex of FeIII to the hydroxamate group of cis-Azo-Gly. Formation
of the 1:3 complex was also confirmed by the molar ratio plot (data not shown). Contrary to our inference,
trans-Azo-Gly also formed the 1:3 complex (λmax=425 nm, ε=2 800), although the three hydroxamate
groups can not get close to each other to form an intramolecular 1:3 complex with a FeIII ion on the
basis of an examination of molecular models. As the intramolecular 1:3 complex is impossible, an
intermolecular cluster structure consisting of two or more molecules is suggested. In order to characterize
the structures of the FeIII complexes, electrospray ionization (ESI) mass spectroscopy was used. The
mass spectrum of the ferric complex with cis-Azo-Gly (Fig. 1a) exhibits a strong sodiated molecular ion
peak (m/z 749.7, 749.2 calculated for [C28H32N9O11Fe·Na]+). Since cis-Azo-Gly is the cis/trans (86/14)
mixture, peaks (m/z 1454.7 and 1475.9, 1453.3 and 1475.3 calculated for [(C28H32N9O11Fe)2·H]+ and
[(C28H32N9O11Fe)2·Na]+, respectively) due to a dimeric complex comprised of two trans-Azo-Glys also
appear. On the other hand, relatively strong peaks (m/z 1453.9 and 1476.1) due to the dimeric complex are
†
Selected data for Azo-Gly: mp 123.5–125°C, IR (KBr)/cm−1: 3600–2700, 1685–1640; δH[(CD3)2SO, 100°C]: 3.13 (s, 9H,
-CH3), 4.04–4.10 (m, 10H, -NCH2-), 4.67 (s, 2H, -OCH2-), 7.20 (d, J=8.8 Hz, 2H, Ar-3H), 7.66 (d, J=8.8 Hz, 2H, Ar-30H), 7.84
(d, J=8.8 Hz, 2H, Ar-2H), 7.89 (d, J=8.8 Hz, 2H, Ar-20H); mass spectrum (FAB, thioglycerol matrix) m/z 674.2522 [(M+H)+;
calcd for C28H36N9O11, 674.2534]. Anal. calcd for (C28H35N9O11·2H2O): C, 47.39; H, 5.54; N, 17.76. Found: C, 47.68; H,
5.73; N, 17.69%.

2413
detected in the spectrum of trans-Azo-Gly (Fig. 1b), whereas no peak corresponding to the monomeric
complex is observed. These mass data support that trans-Azo-Gly forms the 2:2 complex with FeIII.
Furthermore, visible light irradiation of the ferric complex solution of cis-Azo-Gly afforded the same
ESI mass spectrum as that of the trans-Azo-Gly complex. The opposite situation was observed with UV
irradiation of the trans-Azo-Gly complex leading to the monomeric complex. These results indicate that
the structure of monomeric and dimeric complexes is photochemically and reversibly controllable, as
shown in Fig. 2.
Fig. 1. ESI mass spectra of FeIII-cis- (a) and trans- (b) Azo-Gly complexes in DMF
The relative stability of FeIII-complexes with cis-Azo-Gly and trans-Azo-Gly was estimated by the
competition reaction with EDTA.12 Strangely, the stability of FeIII-cis-Azo-Gly was almost equal to that
of trans-Azo-Gly, suggesting that the entropic disadvantage of the 2:2 complex formation with the trans

2414
Fig. 2. Photoregulation of structure of FeIII complex with Azo-Gly
form be cancelled by the enthalpically unstable cis azobenzene structure. However, their stability is below
that of ferrioxamine B, a native trihydroxamate siderophore. Interestingly, FeIII influenced cis–trans
thermal isomerization of Azo-Gly; the thermal isomerization in the presence of FeIII was pronouncedly
slower (half-life 45 h). However, the cis% at the photostationary state remains the same. The first-
order rate constants were determined spectrophotometrically by following the increase in the absorption
band of trans-isomers to be 3.1×10−5 and 4.5×10−5 s−1 at 30°C in the presence and absence of FeIII,
respectively. The fact that FeIII suppresses the rate of thermal cis–trans isomerization also supports the
intramolecular formation of the 1:3 complex of FeIII with the hydroxamate group of cis-Azo-Gly.
In conclusion, the present study demonstrated for the first time that the structure of ferric complex with
the photoresponsive trihydroxamate siderophore containing an azobenzene framework can be regulated
reversibly by irradiating with either visible or UV light. The application of the present findings to the
photoregulation of bacterial growth in which a siderophore is required is currently under way.
Acknowledgements
The authors are grateful to Rika Miyake, Tetsuya Shimada, and Tamaki Nagasawa of the Analytical
Center, Graduate School of Science, Osaka City University, for measurement of mass spectra.
References
1. Feher, G.; Allen, J. P.; Okamura, M.; Rees, D. C. Nature 1989, 339, 111.
2. (a) Burr, D. Curr. Biol. 1999, 11, 90. (b) Rando, R. R. Chem. Biol. 1996, 3, 255. (c) Stryer, L. Am. Rev. Neurosci. 1986, 9,
87.
3. Armitage, J. P. Arch. Microbiol. 1997, 168, 249.
4. (a) Spudich, J. L. Mol. Microbioll. 1998, 28, 1051. (b) Lohmann, D.; Petrak, K. Crit. Rev. Ther. Drug Carrier Syst. 1989, 5,
263.
5. (a) Shinkai, S.; Manabe, O. Top. Curr. Chem. 1984, 121, 67, and references cited therein. (b) Shiga, M.; Takagi, M.; Ueno,
K. Chem. Lett. 1980, 1021. (c) Yamashita, I.; Fujii, M.; Kaneda, T.; Misumi, S. Tetrahedron Lett. 1980, 21, 541. (d) Eichner,
M.; Merz, A. Tetrahedron Lett. 1981, 22, 1315. (e) Blank, M.; Soo, L. M.; Wassermann, N. H.; Erlanger, B. F. Science 1981,
214, 70.
6. Shinmori, H.; Takeuchi, M.; Shinkai, S. J. Chem. Soc., Perkin Trans. 2 1998, 847.
7. Ueno, A.; Yoshimura, H.; Saka, R.; Osa, T. J. Am. Chem. Soc. 1979, 101, 2779.
8. Asanuma, H.; Ito, T.; Yoshida, T.; Liang, X.; Komiyama, M. Angew. Chem., Int. Ed. Engl. 1999, 38, 2393.
9. (a) Kaufman, H.; Vratsanos, S. M.; Erlanger, B. F. Science 1968, 162, 1487. (b) Bieth, J.; Vratsanos, S. M.; Wassermann, N.
H.; Cooper, A. G.; Erlanger, B. F. Biochemistry 1973, 12, 3023.
10. (a) Willner, I. Acc. Chem. Res. 1997, 30, 347. (b) Willner, I.; Rubin, S. Angew. Chem., Int. Ed. Engl. 1966, 35, 367.
11. Sugimoto, H.; Kimura, T.; Inoue, S. J. Am. Chem. Soc. 1999, 121, 2325.
12. Winston, A.; Kirchner, D. J. Am. Chem. Soc. 1978, 100, 597.