A ratiometric fluorescent viscosity sensor

Article abstract of DOI:10.1021/ja056370a

The development of a dual probe that provides ratiometric measurements of fluid viscosity is described. The design is based on coupling of a primary fluorophore with viscosity-independent fluorescence emission (blue unit) with a secondary fluorophore that exhibits viscosity-sensitive fluorescent emission quantum yield (red unit). Excitation of the secondary fluorophore can be achieved via Resonance Energy Transfer. The ratio of the fluorescence emission of these fluorophores provides an accurate, ratiometric measurement of solvent viscosity. Copyright

Full text of DOI:10.1021/ja056370a

Published on Web 12/16/2005
A Ratiometric Fluorescent Viscosity Sensor
Mark A. Haidekker,*,† Thomas P. Brady,‡ Darcy Lichlyter,† and Emmanuel A. Theodorakis*,‡
Department of Biological Engineering, UniVersity of MissourisColumbia, Columbia, Missouri 65211, and
Department of Chemistry and Biochemistry, UniVersity of California, San Diego, La Jolla, California 92093-0358
Received September 15, 2005; E-mail: haidekkerm@missouri.edu; etheodor@ucsd.edu
Variations in fluid viscosity are linked to a variety of functions
and diseases both at the cellular level (e.g., membrane and
cytoplasmic viscosity changes in cell signaling modulation)1 and
at the organismal level (e.g., blood, plasma, or lymphatic fluid
viscosity changes in diabetes, hypertension, infarction, and aging).2
It has been proposed that monitoring of biofluid viscosity could
provide a diagnostic tool for the detection of diseases.3
Since mechanical devices do not provide the spatial and temporal
resolution needed, a new type of fluorescent-based viscosity sensors
was developed.4 These sensors are based on a class of environment-
Figure 1. Structures of fluorescent dyes.
sensitive fluorescent dyes that are characterized by a viscosity-
dependent emission quantum yield.4,5 The chemical structure of
Scheme 1. Synthesis of Fluorescent Dye 4
these dyes contains an electron donor unit (such as a nitrogen atom)
in conjugation with an electron acceptor unit (such as a nitrile).
Upon photoexcitation, the two units can rotate relative to each other
in a manner that is dependent on the viscosity of their environment.
Representative examples of such fluorescent rotors are 9-(dicy-
anoVinyl)julolidine (DCVJ, 1) and 2-cyano-3-(4-dimethylaminophe-
nyl)acrylic acid methyl ester (CMAM, 2) (Figure 1).5 Their
viscosity-dependent fluorescent quantum yield is described by the
Fo¨rster-Hoffmann equation (eq 1).6
then excite the CMAM motif, resulting in a viscosity-dependent
Fluorescent molecular rotors have been used for viscosity studies
that are performed by steady-state fluorescence through emission
intensity measurements. This method suffers, however, from
drawbacks arising from changes of the fluid optical properties and
fluctuations in dye concentrations. An additional disadvantage is
that a calibration curve is needed for the absolute determination of
viscosity.5 As a consequence, changes in fluid properties and dye
concentration may cause erroneous readings.
We hypothesized that a dual dye composed of two distinct
fluorescent units, one providing an internal intensity reference and
the other acting as a viscosity sensor, would create a ratiometric
sensing system, thus overcoming the above disadvantages. Dividing
the sensor emission intensity by the reference emission intensity
would yield a normalized intensity that should not only eliminate
some of the fluid- and concentration-related artifacts but also
provide a means to quantify viscosity by an internal reference. To
test this hypothesis, we synthesized a compound 4 in which the
CMAM motif was coupled with 7-methoxycoumarin-3-carboxylic
acid (MCCA, 3). We chose MCCA as the donor fluorophore in
order to induce excitation of the rotor moiety (CMAM) via
Resonance Energy Transfer (RET).7 We envisioned that, due to its
viscosity-independent quantum yield, MCCA could be used as both
the internal reference and the RET donor. The latter event could
emission of the rotor. The linker was chosen to maintain a distance
between the chromophores in the same range as the Fo¨rster
distance7a to allow considerable energy transfer to the acceptor
combined with measurable donor emission.
The synthesis of compound 4 is depicted in Scheme 1.
Condensation of aldehyde 5 with â-cyanoester 6 produced com-
pound 7 (85%). Deprotection of the primary amine of 7 followed
by coupling with N-Boc-isonipecotic acid (9) afforded amide 10
(85%). Removal of the Boc group of 10 and coupling of the
resulting amine with the succinimidyl ester of MCCA (12) formed
compound 4, a red water-insoluble oil, in 87% yield.
Figure 2 shows the fluorescent fingerprint8 of dye 4 in ethylene
glycol. Peak A corresponds to the fluorescence excitation and
emission of MCCA. RET between the two fluorophores elicits peak
B, while photon-excited emission from the CMAM moiety results
in peak C. Peak B does not occur in a solution containing both of
the unconjugated dyes.
The viscosity sensitivity of dye 4 was tested by measuring its
fluorescent quantum yield in mixtures of ethylene glycol and
glycerol (Figure 3).5,9 Increased glycerol content is known to
increase viscosity with only minimal changes of solvent polarity.10
Glycerol contents of 40, 50, 60, 70, and 80% resulted in viscosities
of 72, 109, 165, 248, and 374 mPa‚s, respectively. As shown in
Figure 3, both the direct rotor emission and the rotor emission
through RET excitation exhibit viscosity sensitivity.
† University of MissourisColumbia.
‡ University of California, San Diego.
9
398
J. AM. CHEM. SOC. 2006, 128, 398-399
10.1021/ja056370a CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
Figure 4. Reduction in the influence of dye concentration using ratiometric
measurements. Blue lines correspond to CMAM emission in mixtures of
ethylene glycol and glycerol (dye concentration 2.5-20 µM). Green lines
correspond to the CMAM emission divided by the MCCA emission.
Ratiometric measurement dramatically reduces variations between samples
despite the high range of concentrations used.
yield, to a secondary fluorophore with a viscosity-sensitive quantum
yield (CMAM). The high quantum yield of MCCA permits the exci-
tation of CMAM via RET while still maintaining sufficient emission
for reference computation. Such a dye can be probed with a single
excitation source with two emission channels, allowing fast ratio-
metric measurements of fluid viscosity. The covalent link between
the dye pair is fundamental to achieve RET as well as to ensure
that local dye concentrations are identical even when preferential
binding sites exist. Viscosity sensitivity covers a wide range of
viscosities from 1 to 400 mPa‚s, and no mechanical device would
be capable of covering a comparable range. The emission ratio
widely eliminates influences of refractive index and dye concentra-
tions, allowing fast and accurate measurements of fluid viscosity.
Figure 2. Fluorescent fingerprint of dye 4. Peak A (λex ) 360 nm and λem
) 402 nm) corresponds to the donor fluorescence peak (MCCA). Peak B
(λex ) 349 nm and λem ) 481 nm) corresponds to the RET, while peak C
(λex ) 449 nm and λem ) 486 nm) is emitted by CMAM.
Acknowledgment. Financial support by the NIH (1R33 018399)
and the UCSD is gratefully acknowledged.
Note Added after ASAP Publication. After this paper was
published ASAP on December 16, 2005, the structure of compound
4 was corrected in Figure 1, the Supporting Information, and the
table of contents. The corrected version was published ASAP
December 20, 2005.
Supporting Information Available: Synthesis and characterization
of compound 4. This material is available free of charge via the Internet
at http://pubs.acs.org.
Figure 3. Emission spectra of probe 4 in mixtures of ethylene glycol and
glycerol. Peaks A, B, and C correspond to MCCA (reference), RET, and
CMAM emissions, respectively. Peaks A and B were acquired at λex
)
360 nm, while peak C was acquired at λex ) 444 nm. Only peaks B and C
References
(1) For selected reviews and monographs, see: (a) Luby-Phelps, K. Int. ReV.
Cytol. 2000, 192, 189-221. (b) Molecular and Cellular Aspects of
Basement Membranes; Rohrbach, D. H., Timpl, R., Eds.; Academic
Press: San Diego, CA, 1993. (c) The Membranes of Cells; Yeagle, P. L.,
Ed.; Academic Press: San Diego, CA, 1993. (d) Membrane Abnormalities
in Sickle Cell Disease and in Other Red Blood Cell Disorders; Ohnishi,
S. T., Ohnishi, T., Eds.; CRC Press: Boca Raton, FL, 1994.
(2) For representative reviews, see: (a) Reinhart, W. H. Biorheology 2001,
38, 203-212. (b) Moriarty, P. M.; Gibson, C. A. CardioVasc. ReV. Rep.
2003, 24, 321-325. (c) Uchimura, I.; Numano, F. Diabetes Frontier 1997,
8, 33-37. (d) Simon, A.; Gariepy, J.; Chironi, G.; Megnien, J.-L.;
Levenson, J. J. Hypertens. 2002, 20, 159-169. (e) Bosman, G. J. C. G.
M.; Bartholomeus, I. G. P.; de Grip, W. J. Gerontology 1991, 37, 95-112.
(3) Harkness, J. Biorheology 1971, 8, 171-193.
(4) (a) Haidekker, M. A.; L’Heureux, N.; Frangos, J. A. Am. J. Physiol. Heart
Circ. Physiol. 2000, 278, H1401-H1406. (b) Haidekker, M. A.; Tsai, A. G.;
Brady, T.; Stevens, H. Y.; Frangos, J. A.; Theodorakis, E. A.; Intaglietta,
M. Am. J. Physiol. Heart Circ. Physiol. 2002, 282, H1609-H1614.
(5) Haidekker, M. A.; Brady, T. P.; Chalian, S. H.; Akers, W.; Lichlyter, D.;
Theodorakis, E. A. Bioorg. Chem. 2004, 32, 274-289.
are viscosity-dependent.
As stipulated by eq 1, the data points of rotor emission intensity
over solvent viscosity, drawn in a double-logarithmic scale, would
lie on a straight line with a slope of x. For this class of dyes, a
slope of x ) 0.6 can be expected.6 Indeed, the rotor emission
maxima exhibited the expected behavior for peaks B (x ) 0.55, R2
) 0.997) and C (x ) 0.52, R2 ) 0.995), while the MCCA reference
peak had a slope of only x ) 0.04 (R2 ) 0.98) (Figure 4). The
minute increase of the reference peak is most likely due to small
changes of refractive index of the solvent. Since refractive index
changes affect both the rotor and the reference emission values,
the ratiometric system eliminates this error.
To verify the ability of dye 4 to compensate for concentration
fluctuations, we prepared mixtures of ethylene glycol and glycerol
with concentrations of 4 ranging from 2.5 to 20 µM. The emission
intensity of the rotor increased not only with increased solvent
viscosity but also with increased dye concentration. Nonetheless,
the ratio of the CMAM peak to the MCCA peak (reference) remains
within a very tight range of values (Figure 4).
(6) Fo¨rster, Th.; Hoffmann, G. Z. Phys. Chem. 1971, 75, 63-76.
(7) (a) Lakowicz, J. R. Energy Transfer. Principles of Fluorescence Spec-
troscopy; Kluwer Academic: New York, 1999. (b) Clegg, R. M. In
Fluorescence Imaging Spectroscopy and Microscopy; Wang, X. F.,
Herman, B., Eds.; John Wiley & Sons: New York, 1996.
(8) Auer, M.; Graf, C.; LaClair, J. J. Angew. Chem., Int. Ed. 2001, 40, 1889-
1892.
(9) Iwaki, T.; Torigoe, C.; Noji, M.; Nakanishi, M. Biochemistry 1993, 32,
7589-7592.
In conclusion, we report herein the first ratiometric viscosity
sensor. The chemical design involves covalent attachment of a pri-
mary fluorophore (MCCA), having a constant emission quantum
(10) Haidekker, M. A.; Brady, T. J.; Lichlyter, D.; Theodorakis, E. A. Bioorg.
Chem. 2005, 33, 415-425.
JA056370A
9
J. AM. CHEM. SOC. VOL. 128, NO. 2, 2006 399