Development of 1,3a,6a-triazapentalene-labeled enterobactin as a fluorescence quenching sensor of iron ion

Article abstract of DOI:10.1016/j.tetlet.2017.04.011

1,3a,6a-Triazapentalene (TAP)-labeled enterobactin was developed as an iron ion sensor. 3-Acetylated-TAP was successfully introduced to the catechol ring of enterobactin, a well-recognized siderophore secreted by various Gram-negative bacteria. The fluorescence of TAP-labeled enterobactin decreased gradually as the amount of Fe³⁺ ion as an additive was increased, and 1.2?equiv of Fe³⁺ ion completely quenched the fluorescence. In clear contrast, when other metal ions were used, the fluorescence of TAP-labeled enterobactin remained even at 5.0?equiv.

Full text of DOI:10.1016/j.tetlet.2017.04.011

Tetrahedron Letters xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Development of 1,3a,6a-triazapentalene-labeled enterobactin as a
fluorescence quenching sensor of iron ion
a
b
c
c
a
Tsukiho Hayashi , Ayumi Osawa , Takehiro Watanabe , Yoshiko Murata , Atsushi Nakayama ,
Kosuke Namba
a
a,
⇑
Graduate School of Pharmaceutical Science, Tokushima University, 1-78 Shomachi, Tokushima 770-8505, Japan
Graduate School of Chemical Science and Engineering, Hokkaido University, Sapporo 060-0810, Japan
Suntory Foundation of Life Sciences, Bioorganic Research Institute, 8-1-1 Seikadai, Seika, Soraku, Kyoto 619-0284, Japan
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
1,3a,6a-Triazapentalene (TAP)-labeled enterobactin was developed as an iron ion sensor. 3-Acetylated-
TAP was successfully introduced to the catechol ring of enterobactin, a well-recognized siderophore
secreted by various Gram-negative bacteria. The fluorescence of TAP-labeled enterobactin decreased
gradually as the amount of Fe ion as an additive was increased, and 1.2 equiv of Fe ion completely
quenched the fluorescence. In clear contrast, when other metal ions were used, the fluorescence of
TAP-labeled enterobactin remained even at 5.0 equiv.
Received 1 March 2017
Revised 3 April 2017
Accepted 4 April 2017
Available online xxxx
3+
3+
Keywords:
Iron ion sensor
Fluorescence quenching
Ó 2017 Elsevier Ltd. All rights reserved.
1
,3a,6a-Triazapentalene
Enterobactin
Siderophore
Iron is an essential element for the vital activity of all living
There have been several examples of iron ion sensors by the
assembly of siderophores and fluorescent organic molecules.6 For
example, Su et al. developed a highly selective iron ion sensor
called FIDFO, which consists of fluorescein and desferrioxamine.6
Thus, we expected that the combination of enterobactin and
fluorescent organic molecules might form an efficient iron sensor
due to enterobactin’s highest affinity for iron ion relative to other
siderophores.
organisms on Earth. However, most organisms suffer from iron-
deficiency stress because most iron exists as poorly water-soluble
trivalent iron salts in the earth’s current environment under condi-
tions of high-oxygen partial pressure. To acquire insoluble iron
efficiently, some kinds of microorganisms have developed an orig-
inal strategy characterized by the synthesis and secretion of side-
rophores as an efficient iron chelator and by the specific uptake
of the resulting iron(III) complex.1 Not surprisingly, siderophores
have been a subject of intensive research due to not only their
physiological importance but also their interesting chemical prop-
erties, such as a high affinity for metal ions and a noteworthy
structural change after complexation.2 Enterobactin (1) is a well-
recognized siderophore produced by various Gram-negative bacte-
ria, and its affinity for ferric ion is one of the highest among known
a
On the other hand, fluorescent organic molecules connecting to
enterobactin adopted the 1,3a,6a-triazapentalene (TAP, 2) as a
compact fluorophore that does not disturb iron complexation or
uptake by bacteria. We recently discovered that the 1,3a,6a-triaza-
pentalene skeleton without an additional fused ring system is a
compact and highly fluorescent chromophore, and that it exhibits
various useful fluorescence properties such as a large Stokes shift
3
7
siderophores (Fig. 1). Thus, the development of enterobactin ana-
and a positive fluorescence solvatochromism. Since the initial dis-
logs as functionalized molecules has been strenuously studied, and
closure of the fluorescence system of TAPs, we have strenuously
studied the application development of TAPs and other groups
4
8
their application to the detection of bacteria and the enhancement
of antibacterial agent activity5 was recently reported. We also
focused on the high iron affinity of enterobactin and planned to
apply it to an iron sensor. Herein, we report a new fluorescent
probe based on enterobactin for the detection of iron ion.
also reported the alternative synthetic method of highly fluores-
9
cent TAPs. Interestingly, the fluorescence wavelength and inten-
sity of 1,3a,6a-triazapentalenes vary widely with the electron
density of 2-substituents. Based on this fluorescence property,
we designed TAP-enterobactin 3, which possesses a TAP system
on the catechol rings of enterobactin. The coordination of catechol
to iron ion should induce a change in the electron state of the
phenyl ring as the 2-substituent of TAP, and this would mean a
⇑
Corresponding author.
E-mail address: namba@tokushima-u.ac.jp (K. Namba).
http://dx.doi.org/10.1016/j.tetlet.2017.04.011
040-4039/Ó 2017 Elsevier Ltd. All rights reserved.
0
Please cite this article in press as: Hayashi T., et al. Tetrahedron Lett. (2017), http://dx.doi.org/10.1016/j.tetlet.2017.04.011

2
T. Hayashi et al. / Tetrahedron Letters xxx (2017) xxx–xxx
Fig. 1. Molecular design of iron sensor.
simultaneous change in the fluorescence of TAP. In addition, the
structural change in TAP-enterobactin 3 induced by complexation
is also expected to change the fluorescence of TAP (Fig. 1).
We first tried to synthesize a catechol moiety possessing a TAP
system, and began this attempt with iodovanillin 4.10 The hydroxyl
groups of 4 were protected by the SEM group after demethylation
to give 5. The CO insertion reaction of 5 in methanol proceeded
smoothly to afford methyl ester 6 in 79% yield, although the neigh-
boring SEM group was removed. Thus, the free hydroxyl group of 6
was protected by SEM group again to give bis-SEM-protected 7,
which was treated with diazophosphonate 811 to afford alkyne 9
in 83% yield. Finally, the method of synthesizing 1,3a,6a-triazapen-
7
talenes developed by our group was applied to alkyne 9, and the
reaction of 9 with azidoditriflate 10 in the presence of a catalytic
amount of copper(I) salt proceeded smoothly to give the desired
TAP-catechol derivative 11 in 80% yield (Scheme 1).
Having prepared TAP-labeled catechol derivative 11, condensa-
tion with the serine trimer 1312 was next examined. After the
hydrolysis of the methyl ester of 11 by treatment with 1.0 equiv
of lithium hydroxide at 85 °C followed by the evaporation of the
solvents, the resulting lithium salt 12 was directly treated with
the serine trimer 13 in the presence of DMT-MM13 as a condensing
reagent, which is available in both alcohol and water solvents, as
reported by Kunishima. However, the desired product 15 was not
obtained, and neither were various other conditions. Thus, we next
tried to construct TAP rings after the condensation of catechol-
alkyne with serine trimer 13. The methyl ester of alkyne 9 was
hydrolyzed under similar conditions to give lithium salt 14. After
the solvents were removed, the condensation of the resulting 14
with 13 by the combined action of DMT-MM and NMMÁHCl in
methanol proceeded to afford the desired condensing product in
Scheme 1. Synthesis of TAP-catechol unit 11.
its poor stability under the conditions of workup and silica gel
column chromatography (Scheme 2). Thus, we should increase
the stability of the TAP system on the catechol ring.
The TAP ring system possessing electron-withdrawing sub-
stituents at the C2-position is basically stable, whereas the elec-
tron-donating derivatives readily decompose under acidic and
oxidative conditions due to the high nucleophilicity at the C3-posi-
tion of TAP. Although the catechol ring system of 15 probably
decreased the stability of TAP, it would be difficult to further
modify the catechol moiety to serve as an iron chelator. Thus, we
next planned to introduce the substituents at the C3-position of
TAP as a protecting group. Since Shi et al. also reported that the
electron-withdrawing groups at the C3-position increase the
5
6% yield. The subsequent TAP-forming reaction of the resulting
alkyne proceeded, and we observed the formation of the desired
1
1
5 by a direct H NMR measurement of the reaction mixture after
9
a
evaporation. However, the isolation of 15 was very difficult due to
stability of 1,3a,6a-triazapentalenes, we attempted to introduce
Please cite this article in press as: Hayashi T., et al. Tetrahedron Lett. (2017), http://dx.doi.org/10.1016/j.tetlet.2017.04.011

T. Hayashi et al. / Tetrahedron Letters xxx (2017) xxx–xxx
3
With TAP-labeled enterobactin 17 in hand, we investigated the
fluorescence change that occurs with the addition of iron(III). To a
solution of 17 in DMSO were added various amounts of Fe(acac)
3
3
+
15
as an Fe
ion source,
and the mixtures were excited at
3
50 nm. The fluorescence maximum of 17 at 525 nm in DMSO
did not shift with the addition of Fe(acac) , whereas the fluores-
3
3+
cence intensity of 17 sensitively responded to the Fe ion. The
fluorescence intensity decreased gradually according to the
3
+
amount of Fe ion, and the fluorescence of 17 completely disap-
peared when 1.2 equiv of Fe(acac) was added (Fig. 2). The forma-
3
3
+
tion of a 1:1 complex of 17 with Fe ion was confirmed by HRMS
spectroscopy. Similar behaviors of fluorescence intensity were also
1
6
observed in DMF and tert-butanol. On the other hand, since 17
did not dissolve in water, its fluorescence behavior in water was
not examined.
Next, the effects of other metals (Al3+, V3+, Ni2+, Cu2+, Ca2+, Li+)
on the fluorescence intensity of 17 were investigated. Various
acetylacetonate complexes were used as a metal source to fix the
conditions. To a 1.0
solutions of M(acac)
50 nm. The fluorescence intensity of 17 at 525 nm without metal
ions was set as 100%, and the relative intensity of fluorescence with
the addition of each equivalent of M(acac) was measured (Fig. 3).
lM solution of 17 in DMSO were added various
n
in DMSO, and the mixtures were excited at
3
n
3+
Interestingly, whereas the addition of 1.0 equiv of Fe ion almost
completely quenched the fluorescence of 17, fluorescence
remained with metals other than iron, although there were some
+
differences. Especially, Li ion remained highly fluorescent with
the addition of 1.0 equiv, and the further additions decreased fluo-
+
+
rescence gradually according to the amount of Li ion. Probably, Li
ion formed a complex with 17 poorly, and fluorescence was still
+
highly observed with the excess amount of Li ion. On the other
Scheme 2. Examination of synthesis of 15.
3
+
hand, the addition of Al ion decreased fluorescence substantially
at the point of 1.0 equiv, although further additions did not further
decrease fluorescence; rather, high levels remained. Since it is
an electron-withdrawing group. After comparing several
substituents at the C3-position of TAP (See Supporting
Information Table S1), we adopted an acetyl group as the best
substituent. Acetylation at the C3-position of TAP was achieved
by treating TAP 11 with acetic anhydride and triethylamine by
reference to Hirobe’s procedure.14 The acetylated 16 was hydro-
lyzed to lithium salt, and subsequent condensation using DMT-
MM with serine trimer 13 in methanol proceeded to give the
desired condensing product in 62% isolated yield. The acetylated
TAP analog was stable enough for purification. Finally, the SEM
groups were deprotected by treatment with anhydrous hydrogen
chloride in 2-propanol to obtain the desired 17 in quantitative
yield (Scheme 3).
3+
3a
already known that Al ion has a high affinity for enterobactin,
3+
1
.0 equiv of Al ion should form a 1:1 complex with 17. Thus,
Al ion halved the fluorescence as a particular property of Al
3+
3+
and/or the half fluorescence was induced by the structural
change in response to complexation. In contrast, the fourth-
2+
3+
2+
2+
period metals equal to iron, such as Ca , V , Cu , and Ni
,
showed sharp declines in fluorescence with the increase of
equivalents. In particular, the fluorescence of transition metals
decreased at 5.0 equiv. Thus, it was found that transition metals
tend to decrease the fluorescence at a rate greater than typical
metals. However, their fluorescence did not completely
disappear, unlike the case with iron ion.
The metals other than iron did not show the complete quench-
ing of the fluorescence of 17 even when 5.0 equiv of metal ions was
3
+
added, whereas 1.2 equiv of Fe ion was enough to quench fluo-
rescence completely. These differences were also observed by the
naked eye. Fig. 4 shows the fluorescent emissions in the presence
À6
of 5.0 equiv of metal ions in 1.0 Â 10 M DMSO solution. The flu-
+
3+
orescence of 17 with Li and Al ions were sufficiently observed
with the results similar to the fluorescence spectral measurement.
2
+
3+
2+
The other metals, such as Ca , V , and Cu , also showed visible
2
+
fluorescence. Although the fluorescence with Ni ion was very
weak, we were able to carefully observe the fluorescence. In clear
contrast, the addition of only 1.2 equiv of Fe ion showed no fluo-
rescence. Therefore, TAP-labeled enterobactin 17 was confirmed to
be a highly selective and sensitive fluorescence-quenching sensor
3
+
3
+
À6
of Fe ion that was effective in 10 M solutions.
In conclusion, we developed a novel iron ion sensor 17 by com-
bining enterobactin as a siderophore and TAP as a fluorescent
organic molecule. Although the synthesis of TAP-labeled enter-
obactin was initially difficult due to the instability of TAP on the
catechol ring, we later found that the 3-acetylated TAP derivative
Scheme 3. Synthesis of TAP-labeled Enterobactin 17.
Please cite this article in press as: Hayashi T., et al. Tetrahedron Lett. (2017), http://dx.doi.org/10.1016/j.tetlet.2017.04.011

4
T. Hayashi et al. / Tetrahedron Letters xxx (2017) xxx–xxx
Fig. 2. Fluorescence spectrum (excited at 350 nm) of 17 with Fe3+ in DMSO.
is grateful to the Naito Foundation, Yamada Foundation, the Kurata
Foundation, and the Uehara Memorial Foundation for support
through a Research Fund.
A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.tetlet.2017.04.
0
11.
References
Fig. 3. Fluorescence intensities of 17 at 525 nm with various metal ions and
equivalents.
1.
(a) Saha R, Saha N, Donofrio RS, Bestervelt LL. J. Basic Microbiol.. 2013;53:303;
(b) Ali SS, Vidhale NN. Int. J. Curr. Microbiol. Appl. Sci.. 2013;2:303;
(
c) Boukhalfa H, Crumbliss AL. BioMetals. 2002;15:325.
2
.
.
Saha M, Sarkar S, Sarkar B, Sharma BK, Bhattacharjee S, Tribedi P. Environ. Sci.
Pollut. Res.. 2016;23:3984.
(a) Loomis LD, Raymond KN. Inorg. Chem.. 1991;30:906;
3
(
b) Perron NR, Brumaghim JL. Cell. Biochem. Biophys.. 2009;53:75.
4.
5.
6.
Lee AA, Chen Y-CS, Ekalestari E, et al. Angew. Chem. Int. Ed.. 2016;55:12338.
Zheng T, Nolan EM. J. Am. Chem. Soc.. 2014;136:9677.
(a) Su B-L, Moniotte N, Nivarlet N, Tian G, Desmet J. Pure Appl. Chem..
2
010;82:2199;
b) Palanche BP, Marmolle F, Abraham MA, Shanzer A, Albrecht-Gray AM. J. Biol.
Inorg. Chem.. 1999;4:188;
c) Delattre F, Cazier-Dennin F, Leleu L, et al. Talanta. 2015;44:451.
Namba K, Osawa A, Ishizaka S, Kitamura N, Tanino K. J. Am. Chem. Soc..
011;133:11466.
(a) Namba K, Mera A, Osawa A, Sakuda E, Kitamura N, Tanino K. Org. Lett..
012;14:5554;
b) Namba K, Osawa A, Nakayama A, et al. Chem. Sci.. 2015;1083:6;
(c) Nakayama A, Nishio S, Otani A, et al. Chem. Pharm. Bull.. 2016;64:830;
(
(
Fig. 4. The pictures of fluorescence with 5.0 equiv of various metal ions. a1.2 equiv
of Fe(acac) was added.
3
7.
2
8.
2
(
could be introduced to the catechol ring of enterobactin. The
synthetic TAP-labeled enterobactin 17 was proven to be a highly
sensitive and iron-specific quenching probe. Since the iron-specific
fluorescence quenching mechanism could not be explained by sim-
ple photoinduced electron transfer (PET) mechanism, the mecha-
nistic details remain to be seen. The computational studies are
currently underway in our laboratory. To prompt the expansion
of 17 to biological studies, such as an iron concentration measure-
ment inside cells, the next subjects of study are how to increase
fluorescence intensity and how to improve solubility in water.
(
(
d) Kamada R, Tano F, Kudoh F, et al. PLoS One. 2016;11:e0160625;
e) Sawada J, Osawa A, Takeuchi T, et al. Bioorg. Med. Chem. Lett.. 2016;26:5765.
9.
(a) Cai R, Wang D, Chen Y, et al. Chem. Commun.. 2014;50:7303;
(b) Verbelen B, Dehaen W. Org. Lett.. 2016;18:6412.
10. Kometani T, Watt DS, Ji T. Tetrahedron Lett.. 1985;26:2043.
1
1. (a) Ohira S. Synth. Commun.. 1989;19:561;
b) Müller S, Liepold B, Roth GJ, Bestmann HJ. Synlett. 1996;6:521.
2. Marinez ER, Salmassian EK, Lau TT, Gutierrez CG. J. Org. Chem.. 1996;61:3548.
(
1
13. Kunishima M, Kawachi C, Hioki K, Terao K, Tani S. Tetrahedron. 2001;57:1551.
1
1
4. Koga H, Hirobe M, Okamoto T. Tetrahedron Lett.. 1978;19:1291.
5. Since 17 was not stable under strongly acidic conditions, Fe(acac)
the complexation under neutral condition.
3
was used for
1
6. The solubility of 17 in ethyl acetate, diethyl ether, dichloromethane, and water
is poor, whereas 17 readily dissolved in DMSO, DMF, n-BuOH and t-BuOH. The
case of methanol needed diluted condition.
Acknowledgments
This work was supported by JSPS KAKENHI Grant Numbers
JP16H01156 in Middle Molecular Strategy and JP16H03292. K.N.
Please cite this article in press as: Hayashi T., et al. Tetrahedron Lett. (2017), http://dx.doi.org/10.1016/j.tetlet.2017.04.011