Oxinobactin, a siderophore analogue to enterobactin involving 8-hydroxyquinoline subunits: Synthesis and iron binding ability

Article abstract of DOI:10.1016/j.bmcl.2008.10.060

Oxinobactin, a siderophore analogue to enterobactin but possessing 8-hydroxyquinoline instead of catechol complexing subunits, has been synthesized starting from l-serine and 8-hydroxyquinoline. Comparative iron binding studies showed that oxinobactin is as effective as enterobactin for the complexation of Fe^III at physiological pH but with improved complexing ability at acidic pH.

Full text of DOI:10.1016/j.bmcl.2008.10.060

Bioorganic & Medicinal Chemistry Letters 18 (2008) 6476–6478
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Oxinobactin, a siderophore analogue to enterobactin involving
8-hydroxyquinoline subunits: Synthesis and iron binding ability
*
Amaury du Moulinet d’Hardemare , Nivine Alnaga, Guy Serratrice, Jean-Louis Pierre
Département de Chimie Moléculaire (UMR 5250), Université Joseph Fourier-Grenoble, 301 rue de la Chimie, BP 53X, 38041 Grenoble cedex 9, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Oxinobactin, a siderophore analogue to enterobactin but possessing 8-hydroxyquinoline instead of cate-
Received 12 September 2008
Revised 13 October 2008
Accepted 13 October 2008
Available online 17 October 2008
chol complexing subunits, has been synthesized starting from L-serine and 8-hydroxyquinoline. Compar-
ative iron binding studies showed that oxinobactin is as effective as enterobactin for the complexation of
FeIII at physiological pH but with improved complexing ability at acidic pH.
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
Siderophore
Enterobactin
8-Hydroxyquinoline
Iron complexes
Iron is vital for organisms, only Lactobacillus and Borelia burgdof-
erie bacteria are known to be iron independent. Due to its very low
solubility (pKs = 38) at neutral pH, FeIII is poorly bio-available. So,
in order to improve FeIII uptake, microorganisms produce low-
molecular weight compounds called siderophores. Siderophores
and most of their synthetic models contain 3 catecholate or
hydroxamate groups, giving stable Fe3+ octahedral complexes.1
However efficient synthetic chelators have been developed involv-
ing other chelating subunits.2 Siderophore-mediated iron transport
in aerobic bacteria involves recognition of the ferri-siderophore by
a receptor on the cell membrane. The recognition pattern is often
sensitive to the chirality of the metal center, which could be in-
duced by the chirality of the ligand3 derived from naturally ami-
no-acids: this is the case with enterobactin (Fig. 1a), the most
efficient siderophore produced by Escherichia coli and Salmonella
Alzheimer disease.6 We have previously shown that chelators
based on 8-hydroxyquinoline exhibit iron complexing abilities of
the same order of magnitude as the catechol-based homologs in
neutral medium, and higher complexing abilities in acidic med-
ium.7 It seemed to us interesting to design a chelator possessing
the most efficient organizing framework, the trilactone derived
from L-serine that also confers the ‘chiral recognition area’, but
coupled to 8-hydroxyquinoline chelating subunits (Fig. 1b). The
aim is to obtain a ligand as strong as enterobactin (which could
eventually be recognized by its protein receptor), with improved
efficiency in acidic medium for targeting other organs (e.g., brain
is more acidic than other organs), and that can form neutral iron
complex (a necessity for crossing the blood brain barrier). We pres-
ent in this preliminary report the synthesis and the FeIII complex-
ing ability of oxinobactin, an analogue of enterobactin involving 8-
hydroxyquinoline instead of the catechol subunits.
typhimurium. It derives from L-serine, forming a macrocyclic trilac-
tone framework tailoring an exceptional preorganization favorable
for the selective complexation of FeIII by its three catechol sub-
units. Interest in synthetic siderophores includes their potential
application as therapeutical iron removal agents (iron overload is
one of the most common metal poisoning) and there is a renewed
interest in this field since Fe3+ has been associated to Cu2+ and Zn2+
in the oxidative stress observed in amyloid plaques in Alzheimer
disease.4,5 Furthermore metal-chelating agents are able to dissolve
Ab-deposits from the amyloid aggregates by removing metal ions,
showing that lipophilic metal chelator like clioquinol (an
8-hydroxyquinoline chelator) have some therapeutic potential in
O
O
O
O
O
O
O
O
O
O
O
O
O
O
NH
OH
O
NH
NH
OH
HN
O
O
NH
HN
O
OH
HO
OH
HO
OH
HO
N
N
N
OH
(a) Enterobactin
(b) Oxinobactin
* Corresponding author. Tel.: +33 47 66 35 705.
Figure 1. (a) Enterobactin (with catechol chelating subunits) and (b) oxinobactin
E-mail address: amaury.d-hardemare@ujf-grenoble.fr (A.M. d’Hardemare).
(with 8-hydroxyquinoline subunits).
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.10.060

A. du Moulinet d’Hardemare et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6476–6478
6477
CO2Me
NH2
O
O
CO2Me
NHTrt
2
O
O
HO
i
ii
HO
O
O
O
O
O
O
O
O
O
1
NH
OH
NH2
+
O
O
O
NH
O
O
HN
H2N
O
NH2
O
O
4
iii
O
O
O
O
OH
i; ii
HO
O
O
NH2
NHTrt
N
NH2
NHTrt
NH2
NHTrt
N
COF
N
3 HCl
N
8
9
OBn
4
3
Scheme 1. Synthesis of the trilactone scaffold 4. Reagents and conditions: (i) TrtCl,
Et3N, CH2Cl2, 79%; (ii) catalyst b, toluene, reflux, 75%; (iii) HCl, EtOH, rt, 95%.
Scheme 3. Coupling step. Reagents and conditions: (i) DIPEA, CH2Cl2, 0 °C, 66%; (ii)
H2, Pd/C, EtOH, 16%.
scribed by Marinez et al.14 Characterization of the product was car-
ried out by 1H NMR, mass spectrometry and by elemental
analysis.15
NHTrt
O
Sn
O
O
Sn
O
O
Sn
O
O
O
N-Trt-β-lactone
from L-serine
The low yield obtained for the final step is explained by the
competitive hydrogenation of the quinoline nucleus as shown by
MS analysis of the crude mixture. Indeed, a similar reaction has
been reported by Shanzer et al. with Hopobactin, a hydroxamate
analogue of enterobactin.16 In order to circumvent this problem
they reported that quantitative debenzylation may be accom-
plished with FeIII, however the ferric complex is formed during
the process. We did not apply this methodology since we needed
the free ligand for competitive study. We also prepared enterobac-
tin following the procedure described in Scheme 3 but starting
from 1,2-dihydroxybenzoic acid.
catalyst a
catalyst b
Figure 2. Formulae of the organotin catalysts a, b and a-lactone of L-serine.
The key compound of the synthesis of enterobactin and oxinob-
-serine (product 4, Scheme
actin is a trilactone ring derived from
L
1). Since the pioneering work of Corey et al.,8 who obtained ente-
robactin in low yield (about 1%), improvement of the synthesis fo-
cused on the macrolactonisation step. The use of an organotin
template by Shanzer et al.9 (Fig. 2, catalyst a) allowed the forma-
tion of the triserine lactone 4 starting from N-trityl-b-lactone of
The complexing abilities of the new ligand were studied by
spectrophotometry competition between enterobactin and oxinob-
actin. The 1:1 iron(III) complex of oxinobactin was prepared in
methanol owing to the poor water solubility of the ligand with
adding diisopropylethylamine (DIPEA) as a base to adjust pH. The
complexation was still achieved in acidic medium (pH ꢀ 3) and
the addition of base has only a slight effect on the UV–visible spec-
L
-serine (Fig. 2) in 26%.
Then in 1997, Gutierrez et al.10 and Raymond et al.11 raised the
yield (85% and 56%, respectively) following the Shanzer approach
but starting from methyl-N-trityl-serinate 2 in place of N-trityl-
b-lactone, which is more tedious to obtain. In our hands, following
these conditions (Scheme 1), reproductive yields of 75% could be
reached only if 0.2 M concentration of 2,2-dibutyl-1,3,2-dioxastan-
nolane12 (Fig. 2, catalyst b) was used as the organotin catalyst in
refluxing toluene.
Removal of the trityl group under acidic conditions gave the tri-
amine lactone 4 as a Tris–HCl salt in quantitative yield. The phenol
group of 8-hydroxyquinoline carboxylic acid 5 (Scheme 2), the che-
lating subunit, was protected in two steps with a benzyl group. Fi-
nally, the carboxylic acid group of compound 7 was activated and
then combined with the trilactone 4.
For this step, we found that better yields and better purity were
obtained with the fluoro acid derivative 8 in place of the more
common chloro derivative. When DAST was used for the fluorina-
tion step, some decomposition occurred. Cyanuryl trifluoride
proved to be more effective for this transformation. This result is
in accordance with the well-established procedures of peptide syn-
theses, and with similar results obtained in the synthesis of Cory-
nebactin.13 Then, removing the benzyl group with Pd/C catalyst
under slightly positive hydrogen pressure afforded oxinobactin 9
in 16% yield (Scheme 3) as a pale yellow hygroscopic compound.
In order to ovoid iron contamination of oxinobactin during the
deprotective step, the glassware was treated as previously de-
trum that exhibits two bands at kmax = 450 nm (
e
= 6600 MÀ1
cmÀ1) and kmax = 590 nm ( = 5000 MÀ1 cmÀ1) (Fig. 3).
e
These features are very close to those of the Tris-(hydroxyquin-
olinate) coordination sphere of FeIII by comparison to the UV–vis-
ible spectrum of Fe-O-TRENSOX complex.17,18 The spectrum of
the ferric complex of enterobactin in methanol at neutral pH is de-
picted in Figure 4 (kmax = 510 nm,
e
= 5100 MÀ1 cmÀ1).
A comparative study of the FeIII complexing abilities of oxinob-
actin and enterobactin was monitored by UV–visible spectrophoto-
metric competition at pH 3.8, 7.1, and 8.9. The equilibrium was
approached from both directions with enterobactin and FeIIIoxin-
obactin or FeIIIenterobactin and oxinobactin as starting reagents.19
The same spectra were obtained. The spectrum recorded at
i
N
CO2H
N
N
CO2Me
OH
6
5
OH
iii
ii
CO2H
N
COF
OBn
7
8
OBn
Scheme 2. Protection and activation of 8-hydroxyquinolinecarboxylic acid.
Reagents and conditions: (i) BF3, MeOH, reflux, 97%; (ii) BnCl, K2CO3, KI, EtOH,
reflux, 95%; then NaOH, MeOH, reflux, then pH 7, 82%; (iii) C3F3N3, DIPEA, CH2Cl2,
0 °C, 75%.
Figure 3. UV–visible absorption spectra of (a) oxinobactine and (b) ferric oxinobactin
in methanol; [Oxinobactin] = 0.95 Â 10À4 M. [FeIII-Oxinobactin] = 0.95 Â 10À4 M,
pH = 7.1.

6478
A. du Moulinet d’Hardemare et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6476–6478
groups (Fig. 6).20 A higher pFe indicates a stronger iron complex.
The pFe values at pH 4, 7.4, and 9 in aqueous solution are 18.8,
29.5, and 31.7, respectively, for O-TRENSOX and 12.8, 29.6, and
35.2, respectively, for TRENCAMS. The results of the spectrophoto-
metric competition of oxinobactin and enterobactin towards FeIII
are consistent with these values.
In summary these results clearly demonstrate that oxinobactin,
an iron chelator possessing the trilactone framework of enterobac-
tin and 8-hydroxyquinoline groups instead of catechol groups, is as
strong as enterobactin in neutral pH but is more effective at lower
pH’s. Various applications of oxinobactin may be expected. The
lipophilicity and neutrality of oxinobactin and its ferric complex
are useful properties that may be of interest in chelation therapy
especially in neurodegenerative diseases (lipophilicity and neutral-
ity are fundamental characteristics to cross the blood brain bar-
rier). Moreover, the iron center in ferrioxinobactin is necessarily
of the same stereochemistry as in ferrienterobactin as it is imposed
by the chirality of the trilactone ligand. We can hypothesize that
ferrioxinobactin would be recognized by the receptor of E. coli or
other pathogenic microorganisms, but its transport properties
would be modified since the ferrioxinobactin is a neutral complex
(ferrienterobactin is trisanionic). Therefore ferrioxinobactin may
act as an antimetabolite. Biological studies are now in progress.
Figure 4. UV–visible absorption spectra of (a) enterobactin and (b) ferric enterobactin
in methanol; [Enterobactin] = 0.98 Â 10À4 M. [FeIII-Enterobactin] = 0.98 Â 10À4 M,
pH = 7.1.
References and notes
1. Telford, J. R.; Raymond, K. N. Siderophores. In Comprehensive Supramolecular
Chemistry; Lehn, J. M., Gokel, G. W., Eds.; Pergamon Press: London, 1996; Vol. 1,
pp 45–266.
2. Albrecht-Gary, A. M.; Crumbliss, A. L. In Metal ions in biological systems; Sigel, A.,
Sigel, H., Eds.; Iron Transport and Storage in Microorganisms, Plants and
Animals; M. Dekker: New York, 1998; Vol. 35, p 239.
3. (a) Raymond, K. N.; Dertz, E. A.; Kim, S. S. Proc. Natl. Acad. Sci. U.S.A. 2003, 100,
3584; (b) Winkelmann, G. In Bioinorganic Chemistry: Transition Metals in Biology
and their Coordination Chemistry; Trautwein, A., Ed.; Wiley, John & Sons, 1997; p
108.
4. Deraeve, C.; Pitié, M.; Mazarguil, H.; Meunier, B. New J. Chem. 2007, 31, 193.
5. Deraeve, C.; Boldron, C.; Maraval, A.; Mazarguil, H.; Gornitzka, H.; Vendier, L.;
Pitié, M.; Meunier, B. Chem. Eur. J. 2008, 14, 682.
6. Ritchie, C.; Bush, A.; Mackinnon, A.; Macfarlane, S.; Mastwyk, M.; McGregor, L.;
Kiers, L.; Cherny, R.; Li, Q.; Tammer, A.; Carrington, D.; Mavros, C.; Volitakis, I.;
Xilinas, M.; Ames, D.; Davis, S.; Beyreuther, K.; Tanzi, R.; Masters, C. Arch.
Neurol. 2003, 60, 1685.
7. Pierre, J.-L.; Baret, P.; Serratrice, G. Curr. Med. Chem. 2003, 10, 1077.
8. Corey, E. J.; Bhattacharyya, S. Tetrahedron Lett. 1977, 45, 3919.
9. Shanzer, A.; Libman, J. J. Chem. Soc., Chem. Commun. 1983, 15, 846.
10. Ramirez, R. J. A.; Karamanukyan, L.; Ortiz, S.; Guterriez, C. G. Tetrahedron Lett.
1997, 38, 749.
11. Meyer, M.; Telford, J. R.; Cohen, S. M.; White, D. J.; Xu, J.; Raymond, K. N. J. Am.
Chem. Soc. 1997, 119, 10093.
12. Deleuze, H.; Maillard, B. J. Organometallic Chem. 1995, 490, C14.
13. Bluhm, M. E.; Kim, S. S.; Dertz, E.; Raymond, K. N. J. Am. Chem. Soc. 2002, 124,
2436.
Figure 5. UV–visible absorption spectrum of a mixture of ferric enterobactine and
oxinobactine in methanol; [FeIII-Enterobactin] = 1.05 Â 10À4 M, [Oxinobactin] =
0.95 Â 10À4 M, pH = 7.1.
pH = 3.8 closely resemble that of FeIIIoxinobactin spectrum. On
raising the pH up to 7.1 (Fig. 5) a shift of the band from 590 nm
to 535 nm was observed; this characteristic showed that both
the species FeIIIoxinobactin and FeIIIenterobactin were still in solu-
tion at this pH. More precisely, analysis of the curve gave at equi-
librium
a
composition of 60% FeIIIenterobactin and 40%
FeIIIoxinobactin. The spectrum recorded at pH = 8.9 was similar
to that of FeIIIenterobactin. These data did not allow to calculate
the formation constant of FeIIIoxinobactin since such a calculation
requires the formation constant of FeIIIenterobactin and the pKa’s
of the two ligands in methanol, that are not known in this solvent.
Nevertheless, these experiments demonstrated that the two li-
gands exhibit similar complexing abilities in neutral medium, and
that oxinobactin is a stronger iron-chelating agent in acidic medium
(i.e., over the pH range 3–7) and a lower one in basic medium. In a
14. Marinez, E. R.; Salmassian, E. K.; Lau, T. T.; Gutierrez, C. G. J. Org. Chem. 1996, 61,
3548.
15. All the glassware was treated according Marinez14 in order to avoid iron
contamination. A solution of compound 8 (0.48 mmol) in 60 mL of freshly
distilled ethanol was subjected to hydrogenation over Pd/C (10%, 100 mg)
under a slightly positive pressure of H2 and debenzylation was monitored by
1H NMR spectroscopy. After completion of the reaction, the catalyst was
removed by filtration and the solvent removed by distillation under vacuum.
The residue then obtained was carefully washed with several portion of freshly
distilled acetonitrile. The remaining solid was then dried under vacuum to
afforded 9 as a pale yellow hygroscopic product (0.046 mmol, 16% yield).
Analytical data of oxinobactin 9. 1H NMR (DMSO, 300 MHz) d: 4.64 (m, 9H, CH–
, CH2); 7.40 (m, 3H, H Ar); 7.53 (m, 3H, H Ar); 7.64 (m, 3H, H Ar); 7.92 (m, 3H, H
Ar); 8.32 (m, 3H, H Ar). MS (ESI) m/z: 773 [MÀH]À 100%. Elemental analysis
(calculated for the trihydrate) C39H36N6O15: calculated C = 56.52; H = 4.38;
N = 10.14; found C = 56.50; H = 4.41; N = 9.44.
previous study, considering the relevant parameter pFe = Àlog[Fe3+
]
calculated for [Fe3+ tot = 10À6 M and [ligand]tot = 10À5 M, we have
]
compared the efficiency of two iron tripodal chelators, O-TRENSOX
based on 8-hydroxyquinoline and TRENCAMS based on catechol
16. Weizman, H.; Shanzer, A. Chem. Commun. 2000, 2013.
17. Baret, P.; Béguin, C.; Boukhalfa, H.; Caris, C.; Laulhère, J.-P.; Pierre, J.-L.;
Serratrice, G. J. Am. Chem. Soc. 1995, 117, 9760.
18. Serratrice, G.; Boukhalfa, H.; Béguin, C.; Baret, P.; Caris, C.; Pierre, J.-L. Inorg.
Chem. 1997, 36, 3898.
19. Spectra were recorded 48 h after mixing to ensure equilibrium condition. The
complexes are stable during this time.
R =
R =
R
O
OH
OH
N
OH
O
NH
N
H
R
N
SO3Na
O-TRENSOX
SO3Na
TRENCAMS
HN
R
O
20. Albrecht-Gary, A. M.; Blanc, S.; Biaso, F.; Thomas, F.; Baret, P.; Gellon, G.; Pierre,
J. L.; Serratrice, G. Eur. J. Inorg. Chem. 2003, 14, 2596.
Figure 6. Formulae of O-TRENSOX and TRENCAMS.