Fluorescent metal ion indicators based on benzoannelated crown systems: A green fluorescent indicator for intracellular sodium ions

Article abstract of DOI:10.1016/j.bmcl.2005.02.017

The synthesis and metal binding properties of cation-sensitive fluorescent indicators intended for biological applications are described. The increase of the crown ether ring size enhances the affinity for larger cations, but weakens the fluorescent response and selectivity. A compound having a 15-crown-5 chelator directly attached to a 2,7-difluoroxanthenone fluorophore loads into live cells and responds to sodium ion concentration changes with large fluorescence increases in the visible wavelength range.

Full text of DOI:10.1016/j.bmcl.2005.02.017

Bioorganic & Medicinal Chemistry Letters 15 (2005) 1851–1855
Fluorescent metal ion indicators based on benzoannelated crown
systems: a green fluorescent indicator for intracellular sodium ions
Vladimir V. Martin, Anca Rothe and Kyle R. Gee*
Molecular Probes Invitrogen Detection Technologies, 29851 Willow Creek Road, Eugene, OR 97402, USA
Received 30 November 2004; revised 4 February 2005; accepted 7 February 2005
Abstract—The synthesis and metal binding properties of cation-sensitive fluorescent indicators intended for biological applications
are described. The increase of the crown ether ring size enhances the affinity for larger cations, but weakens the fluorescent response
and selectivity. A compound having a 15-crown-5 chelator directly attached to a 2,7-difluoroxanthenone fluorophore loads into live
cells and responds to sodium ion concentration changes with large fluorescence increases in the visible wavelength range.
Ó 2005 Elsevier Ltd. All rights reserved.
Ion channels controlling the flow of monovalent sodium
and potassium cations form the signal transduction net-
works of intra- and intercellular communication sys-
tems.1 The study of ion concentrations, gradients, and
currents is integral to most areas of cellular research,2
drug screening,3 and clinical diagnostics.4 Fluores-
cence-based reporting enjoys a centerpiece role in bio-
logical detection techniques by providing quantitative,
temporal, and spatial information.5 Several fluorescent
reagents have been developed for the quantification
and imaging of biologically important metal ions, such
as magnesium,6 zinc,7 and especially calcium.8 However,
only a few fluorescent reagents for the detection of so-
dium and potassium ions inside live cells have been re-
ported and only two of them, SBFI and Sodium
Green, are available commercially.9,10 The disadvantage
of SBFI is that the excitation wavelength is in the UV
range, while visible wavelength-excited Sodium Green
displays only a modest fluorescence increase in response
to Na+ binding. Also, because both molecules include
two fluorophores, they are bulky and exhibit difficulties
loading into live cells. Recently we reported the syn-
thesis of new fluorescent Na+ indicators of the general
formula 1 based on a benzoannelated crown ether sys-
tem.11 Sodium indicators of this series respond well to
Na+ binding in vitro; however only the tetramethylrhod-
amine derivative 1 (Y = COOCH3; Fl = TMR), aka
CoroNa Red is capable of loading into live cells where
it compartmentalizes in mitochondria.11,12 We hypothe-
sized that substitution of the second aniline fragment,
which does not participate in fluorescent reporting, by
an ethyleneoxy bridge (i.e., transforming compounds 1
into a new series 2) could decrease the moleculesÕ bulk
and improve loading properties and binding affinities.
O
O
O
O
Y
N
N
Y
Y
N
O
O
Y
N
O
O
O
O
O
Fl
Fl
1
2
Fl
Fl = fluorophore
Y= sidearm substituent
Keywords: Fluorescence; Sodium indicator; Fluoroionophore.
*
Corresponding author. Tel.: +1 541 335 0366; fax: +1 541 344 6504; e-mail: kyle.gee@probes.com
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.02.017

1852
V. V. Martin et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1851–1855
O
O
O
O
-N
O
ii
NH2
HN
O
CH3OOCCH2
O
O
OH
i
n
n
3
4a-c
5a-c
iii
iv
O
O
O
O
CH3OCH2CO-N
Y
N
O
O
iv
v
O
O
n
6
CHO
4a,5a,7a n = 0. 4b,5b,7b,8 n = 1. 4c,5c,7c n = 2.
7a-c Y = COOCH3. 8 Y = CH2OCH3.
7a-c, 8
Scheme 1. Reagents and conditions: (i) TsO(CH2CH2O)n(CH2CH2O)3Ts, CsF, MeCN (25–62%); (ii) BrCH2COOCH3, DIEA, MeCN (52–65%); (iii)
CH3OCH2COCl, DIEA, CHCl3 (55%); (iv) POCl3, DMF (55–78%); (v) BH3, THF (77%).
We also wanted to investigate the effect of the size of the
crown ring on binding of different monovalent cations.
cluded acylation with methoxyacetyl chloride, followed
by borane reduction of the resulting amide 6 and intro-
duction of the formyl group.
The synthesis of compounds 2 started with N,O-dialkyl-
ation of 2-aminophenol 3 with polyethyleneglycol
ditosylates (Scheme 1) in the presence of CsF.13,14 Com-
pounds 4a–c were converted into aldehydes 7a–c by
alkylation with methyl bromoacetate, followed by Vils-
meier formylation of the intermediates 5a–c. As a result,
we synthesized a series of key benzaldehydes containing
the 12-crown-4 (7a), 15-crown-5 (7b), or 18-crown-6 (7c)
chelators. Aldehyde 8, having a 15-crown-5 moiety and
methoxyethyl sidearm (Y = CH2OCH3), was prepared
from compound 5b in a three step sequence, which in-
The aldehyde group in compounds 7, 8 was transformed
into several fluorophores by the procedures developed
for the synthesis of fluorescent Ca2+ indicators15 and
also used earlier in preparation of Na+ indicators 111
(Scheme 2).
Another 15-crown-5 compound 17 having an amide
linkage between the sensor and crown moiety, similar
to that of the calcium indicator Oregon Green 488 BAP-
TA 488,15 was prepared starting with acylation of amino
ii
i
7a-c, 8
9a-d
9e
9a-c, 10a, 11a, 12 Y = COOCH3.
O
O
iii
9b
9d, 10b, 11b Y = CH2OCH3.
9e Y = COOH.
ii
vi
ix
iv
v
O
Y
N
9a n = 0.
10a,b
9b, 9e, 10a-b, 11a-b, 12 n = 1.
9d n =2.
7b, 8
O
n
11a,b
12
vii
viii
Fl
7b
9 -12
10 Fl =
9 Fl =
12 Fl =
11 Fl =
CH3
CH3
H3C
F
F
NH
CH3
H3C
HO
O
O
N
O
N
N
N
Cl
B
CH3
H3C
H3C
F
F
COOCH3
INDO
Oregon Green
BODIPY
Tetramethylrhodamine (TMR)
Scheme 2. Reagents and conditions: (i) 4-fluororesorcinol, CH3SO3H (75–90%); (ii) chloranil, MeOH/CHCl3 (25–39%); (iii) KOH/H2O (59%); (iv) 3-
dimethylaminophenol, EtCOOH/TsOH (60%); (v) 4-methoxycarbonyl-2-nitrobenzylphosphorane/K2CO3, DMF (62–71%); (vi) P(OEt)3 (64–75%);
(vii) 2,4-dimethylpyrrole/TFA; (viii) DDQ; (ix) Et2OBF3 (30% in three steps).

V. V. Martin et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1851–1855
1853
ii
Q-NO2
13
Q-NH2
i
14
iv
Q-H
5b
iii
v
Q-N=N
SO3H
Q =
15
CONH-Q
O
CONH-Q
-
CH3OOCCH2 N
O
vi
CO
O
COOH
O
O
F
F
F
F
AcO
O
16
OAc
HO
O
17
O
Scheme 3. Reagents and condition: (i) HNO3/Ac2O (57%); (ii) H2/Pd (62%); (iii) sulfanilic acid/NaNO2, AcOH (8%); (iv) Na2S2O4 (85%); (v) 3,6-
diacetyl-50-carboxy-2,7-difluorofluorescein/(COCl)2, CH2Cl2/Et3N (63%); (vi) Et3N/H2O (48%).
crown 14 with a protected Oregon Green acid chloride,
followed by hydrolytic deprotection with aqueous Et3N.
The key aminocrown ether 14 was prepared by the
reduction of either nitro derivative 13 or azo dye 15;
the former reaction produced a mixture which was easier
to purify (Scheme 3).
cantly. Surprisingly, reduction of the crown size
from 15 to 12 (compound 9a vs 9b) reduces both
binding and fluorescence response to every tested
cation, including the smaller Li+.
(b) Sodium binding affinities of the 15-crown-5 com-
pounds are similar for the compounds having neu-
tral fluorophores (9b, 11a, and 12), while decreasing
for the positively charged tetramethylrhodamine
derivative 10a. This effect can be attributed to the
reduced electron density on the aniline nitrogen of
the indicator group. The large fluorescence increase
as a function of Na+ binding to 9b (aka CoroNa
Green) is shown in Figure 1. Compounds 11a–b
having Indo-type fluorophores do not display a
pronounced ratiometric response, as in the earlier
series 1,11 but behave as on-off switches.
The in vitro study of ion binding was conducted by fluo-
rescence spectroscopy using a series of pH 7.0 buffer
solutions containing 0–1000 mM concentration of
monovalent Li+, Na+, K+, and Rb+ cations in 50 mM
MOPS at 20 °C (Fig. 1). Several conclusions can be
drawn from the results, listed in Table 1.
(a) Increase of the crown ring size from 15 to 18 (com-
pound 9c vs 9b) increases binding affinity of the lar-
ger K+ and Rb+ cations; however the magnitude of
the fluorescent response to binding drops signifi-
(c) The change in sidearm substituent
Y from
COOCH3 to CH2OCH3 (compounds 9b, 10a, and
11a vs 9d, 10b, and 11b) results in a small increase
in binding affinity, possibly because of the replace-
ment of an electron-withdrawing carbonyl group by
methylene group; however the magnitude of the
fluorescent response to ion binding decreases. A
similar effect is observed for compound 9e, which
should be negatively charged at pH 7.
1000 mM
Ex = 492 nm
mM
500
300mM
(d) Selectivity of Na+ versus K+ binding is about 4 for
the CoroNa Green (9b), and decreases to less than 2
for tetramethylrhodamine (10b) or BODIPY (12)
derivatives. This is in line with other crown ether
indicators, including our previous series 1. Com-
pound 9b displays a binding Kd of about 100 mM
when titrated in the presence of 100 mM KCl mak-
ing it possible to use in physiological conditions. As
for selectivity over Ca2+ binding, 9b is impervious
to this important divalent cation at physiologically
relevant levels. A very small fluorescence increase
(S = 1.3) is observed upon addition of Ca2+, but
not until the [Ca2+] reaches several hundred micro-
molar (Kd 470 lM).
mM
200
mM
100
80mM
50 mM
20 mM
mM
10
5
0
mM
mM
560
540
Wavelength (nm)
500
520
580
600
Figure 1. Fluorescence response of compound 9b to increasing [Na+]
in 50 mM MOPS (pH 7 adjusted with tetramethylammonium
hydroxide).
For in cyto studies Jurkat cells were loaded with 10 lM
dye for 1 h at 37 °C. Cells were washed and resuspended

1854
V. V. Martin et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1851–1855
Table 1. Cation binding data
Compound
Fluorescence (nm)
Cation binding, Kd (mM)
Excit.
Emiss.
Li+
Na+
K+
Rb+
Kd
Sa
Kd
S
Kd
S
Kd
S
9a
9b
493
492
492
494
491
546
546
351
351
498
494
517
516
516
516
518
576
576
444
444
510
522
438
142
2.8
6.8
226
82
19.7
28.5
1.5
5.9
3.6
41
978
291
11
9.7
376
319
25
2.9
2.7
1.5
6.9
2.2
2.8
b
9c
9d
15
78.8
269
9e
20
163
150
89
10a
10b
11a
11b
12
254
10.6
12
25
b
89
2
100
59
7.3
3.9
170
144
2.9
2.7
17
65
3.1
157
1.9
a S = I[bound]/I[free] fluorescent response to a cation binding (I = fluorescence intensity).
b Low sensitivity (outside the error margin).
O
O
O
O
O
Y
N
O
Y
N
O
O
i
F
F
F
F
HO
O
O
H3COCOH2CO
O
O
9b,d
18a,b
9b, 18a Y = COOCH3. 9d, 18b Y = CH2OCH3.
Scheme 4. Reagents and condition: (i) BrCH2OCOH3/DIEA, DMF (72–86%).
in sodium gluconate in the presence of 20 lM monensin
for 30 min. Fluorescence was measured by flow cytome-
try. The initial results showed only poor loading with
every 15-crown-5 derivative 9–12. The cell loading in-
creased dramatically when compound 9b was converted
into acetoxymethyl ether 18a (Scheme 4).
In side-by-side experiments compound 18a loads into
20% of a cell population compared to 6% for 9b (Fig.
2) and 3% for Sodium Green tetraacetate. Also, 18a dis-
played a larger fluorescence response to sodium influx,
compared with Sodium Green tetraacetate (Fig. 2), that
is, 64x fluorescence increases were regularly observed.16
Surprisingly, the structurally similar compound 18b
loaded into live cells poorly. A report appeared recently
describing a somewhat similar green fluorescent sodium
ion indicator.17 However, the binding affinity and fluo-
rescence increases observed upon Na+ binding were low-
er than those obtained for 9b; additionally no live cell
data were presented.
CoroNa Green
Sodium Green
100
0
20
50
145
In conclusion, a novel series of fluorescent sodium ion
indicators has been synthesized and evaluated for ion
binding in aqueous solutions and in live cells. The 15-
crown-5 derivative 18a (aka CoroNa Green AM16) per-
formed well as a fluorescent indicator for sodium ions in
cyto.
Sodium (mM)
Figure 2. Fluorescence intensity of live Jurkat cells loaded with 10 lM
18a (CoroNa Green AM) or 10 lM Sodium Green tetraacetate in the
presence of 20 lM monensin. Fluorescence was measured by flow
cytometry.

V. V. Martin et al. / Bioorg. Med. Chem. Lett. 15 (2005) 1851–1855
1855
8. (a) Minta, A.; Kao, J. P. Y.; Tsien, R. Y. J. Biol. Chem.
1989, 264, 8171; (b) Martin, V. V.; Beierlein, M.; Morgan,
J. L.; Rothe, A.; Gee, K. R. Cell Calcium 2004, 36,
509.
References and notes
1. Hille, B. Ionic Channels of Excitable Membranes, 3rd ed.;
Sunderland, MA: Sinauer Associates, 2001.
9. Minta, A.; Tsien, R. Y. J. Biol. Chem. 1989, 264, 19449.
10. Kuhn, M., et al. U.S. Patent 5405975, 1995. Chem. Abstr.
1995, 122, 50761.
11. Martin, V. V.; Rothe, A.; Diwu, Z.; Gee, K. R. Bioorg.
Med. Chem. Lett. 2004, 14, 5313.
12. Bioprobes; Molecular Probes: Eugene, OR, 2001; 36, p 8.
13. Lockhart, J. C. J. Chem. Soc., Perkin Trans. 1 1973,
577.
2. Receptor and Ion-Channel Trafficking; Moss, S. J., Henley,
J., Eds.; New York: Oxford University Press, 2002.
3. Xu, J.; Wang, X.; Ensign, B.; Li, M.; Wu, L.; Guia, A.;
Xu, J. Drug Discov. Today 2001, 6, 1278.
4. (a) Ashcroft, F. M. Ion Channels and Disease: Channelo-
pathies; New York: Academic, 2000; (b) Shieh, C.-C.;
Coghlan, M.; Sullivan, J. P.; Gopalakrishnan, M. Phar-
macol. Rev. 2000, 52, 557.
14. Navarro, P.; Rodrigues-Franco, M. I.; Foces-Foces, C.;
Cano, F.; Samet, A. J. Org. Chem. 1989, 54, 1391.
15. Haugland, R. P. Handbook of Fluorescent Probes and
Research Products, 9th ed.; Molecular Probes: Eugene,
OR, 2002; Chapter 20.
16. Bioprobes; Molecular Probes: Eugene, OR, 2004; 46, p 30.
17. Kenmoku, S.; Urano, Y.; Kanda, K.; Kojima, H.;
Kikuchi, K.; Nagano, T. Tetrahedron 2004, 60, 11067.
5. Lakowicz, J. R. Principles of Fluorescence Spectroscopy,
2nd ed.; New York: Plenum Publishing, 1999.
6. Ishijima, S.; Uchibori, A.; Takagi, H.; Maki, R.; Ohnishi,
M. Arch. Biochem. Biophys. 2003, 412, 126.
7. (a) Chang, C. J.; Jaworski, J.; Nolan, E. M.; Sheng, M.;
Lippard, S. J. Proc. Natl. Acad. Sci. U.S.A. 2004, 101,
1129; (b) Gee, K. R.; Zhou, Z. L.; Ton-That, D.; Sensi, S.
L.; Weiss, J. H. Cell Calcium 2002, 31, 245.