and 768 nm, with the corresponding emission maxima at
782 and 784 nm, respectively. Even though the molar
extinction coefficients of 1 (ꢀ ) 176 000 M-1 cm-1) and 2
(ꢀ ) 157 000 M-1 cm-1) decreased compared to control
molecule 15 (ꢀ ) 192 000 M-1 cm-1) due to the “dye
dilution” effect,14 both of these new compounds demonstrated
much higher quantum yields than 15 (0.043, Table 1). The
Table 1. Photophysical Properties of 1, 2, and 15
λmax,abs (nm)
ꢀ (M-1 cm-1
)
λmax, ema (nm)b
Φfb
1c
766
768
766
779
794
176 000
157 000
192 000
191 000
218 000
782
784
783
802
816
0.186
0.113
0.043
0.041
0.13
2c
15c
ICGc
ICGd
Figure 2. pH-dependent fluorescence intensity profile of 2 and 1
(inset) in 0.1 M sodium phosphate buffer. The fluorescence
intensities of 2 and 1 (0.6 µM) were determined at 784 and 782
nm with excitation at 768 and 766 nm, respectively.
a The fluorophores were excited at their maximum absorbances. b Quan-
tum yields were determined by comparison with indocyanine green (ICG,
Figure 1) in DMSO (Q.Y. ) 0.13), at 298 K. c Measured in PBS buffer,
pH 7.4 (<0.005% methanol as cosolvent). d Measured in DMSO.
3-fold intensity enhancement compared to pH 7.2. This pH-
dependent fluorescence was reversible in the pH range 3.0-
7.0 for at least three cycles. Compound 2 may prove useful
in future applications, where acidic environments, such as
the lysosomal lumen (pH ≈ 5.0)3,4 and/or the tumor mi-
croenvironment characterized by extracellular pH values of
6.4-6.8,18 can enhance its NIR fluorescence. Intermolecular
electrostatic repulsions arising from protonation of both
secondary and primary amines in glucosamine may alleviate
self-quenching between cyanine fluorophores and hence
increase the fluorescence.
Previous studies reported that covalently binding saccha-
rides to initially toxic polymers or small molecules can
substantially decrease the toxicity of the resulting conju-
gates.19 Here, we tested the cytotoxicity of 1 and 2 relative
to 15 in four human mammary epithelial cell (HMEC) lines,
MCF-12A, MCF-7, MDA-MB-231, and MDA-MB-435,
which represent different stages of malignancy, by MTT
assay.20 In all four HMEC lines, 1 and 2 exhibited signifi-
cantly lower cytotoxicity than 15. No cells survived following
treatment with 15 at concentrations above 250 µM. Com-
pound 1 demonstrated minimal cytotoxicity, resulting in cell
viabilities of 95%-100% in MCF-12A and MCF-7 cells even
at the highest concentration (2.0 mM) (Figure 3). Although
1 and 2 exhibited minimal toxicity in MCF-7 and MCF-
12A cells within the tested concentration range, both
compounds caused significant cytotoxicity in the highly
fluorescence quantum yields (Φf) of 1 and 2 were 0.186 and
0.113, respectively, which correspond to a 4- and 3-fold
increase compared to the control molecule. These high
quantum yields of 1 and 2 can be explained by the presence
of multiple hydrophilic groups in the glucosamine moiety,
which may not only improve the hydrophilicity of these
conjugates but also disrupt the π-π interaction between the
hydrophobic carbocyanine cores.15 Reduced quantum yields
of carbocyanine dyes in aqueous environments have previ-
ously been reported as a consequence of their self-aggrega-
tion.15,16 To test this hypothesis, we studied the concentration
dependence of the absorbances of 1, 2, and 15. An additional
broad, blue-shifted band centered at 670 nm increased with
concentrations of 15 (Supporting Information, Figure S2),
which most likely corresponded to H aggregates.14 In
contrast, no additional bands were found in the absorption
spectra of 1 and 2, not even at concentrations above 50 µM.
These results indicate that 1 and 2 have decreased aggrega-
tion tendencies in aqueous solution compared to that of
control molecule 15.
To date, very few pH-dependent NIR dyes have been
reported.8b,17 pH effects on the fluorescence of 1 and 2 were
investigated. Similar to control molecule 15, the emission
intensity of 1 decreased with acidification (Figure 2).
Interestingly, compound 2 exhibited a quite unique pH-
dependent fluorescence behavior. Between pH 11.0 and 7.2
(slope a, Figure 2), its fluorescence decreased significantly
with decreasing pH, reaching a minimum at pH 7.2. Upon
further acidification, the fluorescence increased inversely
(slope b, Figure 2), reaching a maximum at pH 3.0 with a
(18) Stubbs, M.; McSheehy, P. M.; Griffiths, C. L. Mol. Med. Today
2000, 6, 15-19.
(19) (a) Metzke, M.; O’Connor, N.; Maiti, S.; Nelson, E.; Guan, Z.
Angew. Chem., Int. Ed. 2005, 44, 6529-6533. (b) Queiroz, E. F.; Roblot,
F.; Duret, P.; Figade`re, B.; Gouyette, A.; Lapre´vote, O.; Serani, L.;
Hocquemiller, R. J. Med. Chem. 2000, 43, 1604-1610.
(20) The MTT cell proliferation assay determines the ability of viable
cells to reduce yellow [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide] (MTT) to blue-colored formazan crystals by mitochondrial
enzymes. The concentration of formazan crystals can be determined
spectrophotometrically when dissolved in an organic solvent. For a detailed
description of this assay, see: Cetin, Y.; Bullerman, L. B. J. Agric. Food
Chem. 2005, 53, 6558-6563.
(13) Gabor, P.; Narasimhachari, N.; Lucjan, S.; Lyle, R. M.; Malgorzata,
L. US Patent 5,571,388, November 5, 1996.
(14) Ye, Y. P.; Li, W. P.; Anderson, C. J.; Kao, J.; Nikiforovich, G. V.;
Achilefu, S. J. Am. Chem. Soc. 2003, 125, 7766-7767.
(15) (a) Khairutdinov, R. F.; Serpone, N. J. Phys. Chem. B 1997, 101,
2602-2610. (b) Egorov, V. V. J. Chem. Phys. 2002, 116, 3090-3103.
(16) Zhang, Z. R.; Achilefu, S. Org. Lett. 2004, 6, 2067-2070.
(17) Zhang, Z. R.; Achilefu, S. Chem. Commun. 2005, 5887-5889.
Org. Lett., Vol. 8, No. 17, 2006
3625