J. Phys. Chem. B 2008, 112, 2829-2836
2829
Temperature-Dependent Fluorescence Lifetime of a Fluorescent Polymeric Thermometer, Poly(N-isopropylacrylamide), Labeled by Polarity and Hydrogen Bonding Sensitive 4-Sulfamoyl-7-aminobenzofurazan Chie Gota,† Seiichi Uchiyama,*,† Toshitada Yoshihara,‡ Seiji Tobita,‡ and Tomohiko Ohwada† Graduate School of Pharmaceutical Sciences, The UniVersity of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan, and Graduate School of Engineering, Gunma UniVersity, 1-5-1 Tenjin-cho, Kiryu, Gunma 376-8515, Japan ReceiVed: October 8, 2007; In Final Form: December 13, 2007
Fluorescent molecular thermometers showing temperature-dependent fluorescence lifetimes enable thermal mapping of small spaces such as a microchannel and a living cell. We report the temperature-dependent fluorescence lifetimes of poly(NIPAM-co-DBD-AA), which is a random copolymer of N-isopropylacrylamide (NIPAM) and an environment-sensitive fluorescent monomer (DBD-AA) containing a 4-sulfamoyl-7aminobenzofurazan structure. The average fluorescence lifetime of poly(NIPAM-co-DBD-AA) in aqueous solution increased from 4.22 to 14.1 ns with increasing temperature from 30 to 35 °C. This drastic change in fluorescence lifetime (27% increase per 1 °C) is the sharpest ever reported. Concentration independency, one of the advantages of fluorescence lifetime measurements, was seen in average fluorescence lifetime (13.7 ( 0.18 ns) of poly(NIPAM-co-DBD-AA) at 33 °C over a wide concentration range (0.005-1 w/v%). With increasing temperature, polyNIPAM units in poly(NIPAM-co-DBD-AA) change their structure from an extended form to a globular form, providing apolar and aprotic environments to the fluorescent DBD-AA units. Consequently, the environment-sensitive DBD-AA units translate the local environmental changes into the extension of the fluorescence lifetime. This role of the DBD-AA units was revealed by a study of solvent effects on fluorescence lifetime of a model environment-sensitive fluorophore.
1. Introduction Fluorescent molecular thermometers showing temperaturedependent fluorescence properties are useful for temperature measurements in small spaces, since they function at the molecular scale and can be detected by fluorometry.1,2 Until now, various kinds of fluorescent molecular thermometers have been applied to thermal imaging of a microchannel,3,4 a microreactor,5 and even a plasma membrane.6,7 Fluorescence intensity of fluorescent molecular thermometers is a typical parameter, which varies with a change in temperature. In spite of its popularity due to simplicity, fluorescence intensity is seriously affected by fluctuations in experimental conditions (e.g., excitation light intensity, solution turbidity, and concentration of fluorescent molecular thermometers8,9). In fact, the intensity of an excitation light is inconstant in a microchannel because the refractive index of a fluid flow is always varying. In a living cell, the concentration of a fluorophore is not fixed because the cellular volume is continuously changing. Fluorescence lifetime is an effective parameter for temperature measurements in such unstable conditions because it is independent of the excitation light intensity, the solution turbidity, and the concentration of fluorescent molecular thermometers.8,9 Thus, fluorescence lifetime measurements with fluorescent molecular thermometers are capable of measuring accurate temperature without a complicated process of calibration. Additionally, recent improvements in a detection equipment * To whom correspondence should be addressed. E-mail: seiichi@ mol.f.u-tokyo.ac.jp. Phone: +81 3 5841 4768. † The University of Tokyo. ‡ Gunma University.
have enhanced the usefulness of fluorescence lifetime imaging (FLIM).10-14 In the literature, we can find a report on threedimensional temperature mapping of a microchannel by FLIM with rhodamine B, which is a fluorophore moderately sensitive to temperature variation.15 Thus, the fluorescent molecular thermometers that show fluorescence lifetimes highly responsive to temperature variation are expected to be powerful tools for temperature measurements in small spaces. In the present study, we investigated the fluorescence lifetime of a fluorescent polymeric thermometer, poly(NIPAM-co-DBDAA) (Figure 1a), in aqueous solution at various temperatures. It is noteworthy that fluorescent polymeric thermometers based on a random copolymer of NIPAM and a fluorescent benzofurazan have shown high sensitivity to temperature variation when fluorescence intensity is adopted as a temperaturedependent parameter.16-18 Their maximum excitation and emission wavelengths in the visible light region (i.e., around 460 and 550 nm, respectively) and high brightness (i.e., molar absorption coefficient >104 M-1 cm-1 and fluorescence quantum yield Φf > 0.10) meet the requirements in biological applications: these features can avoid a background noise from biomolecules in temperature measurements. The functions of the fluorescent polymeric thermometers derive from their heatinduced structural change from a hydrated open form to a dehydrated globular form, which results in an enhancement of fluorescence signal. Since fluorescence lifetime is another photophysical parameter correlating with the fluorescence efficiency, poly(NIPAM-co-DBD-AA) is a good candidate for a fluorescent molecular thermometer with fluorescence lifetime highly sensitive to temperature variation. In addition to the
10.1021/jp709810g CCC: $40.75 © 2008 American Chemical Society Published on Web 02/16/2008
2830 J. Phys. Chem. B, Vol. 112, No. 10, 2008
Figure 1. Chemical structures of studied compounds. NIPAM, N-isopropylacrylamide; DBD-AA, N-{2-[(7-N,N-dimethylaminosulfonyl)-2,1,3-benzoxadiazol-4-yl](methyl)amino}ethyl-N-methylacrylamide; DBD-IA, N,2-dimethyl-N-(2-{methyl[7-(dimethylsulfamoyl)2,1,3-benzoxadiazol-4-yl]amino}ethyl)propanamide; DBThD-IA, N,2dimethyl-N-(2-{methyl[7-(dimethylsulfamoyl)-2,1,3-benzothiadiazol-4yl]amino}ethyl)propanamide; BD-DA, 5-N,N-dimethylamino-2,1,3benzoxadiazole.
sensitivity, concentration independency was ascertained for fluorescence lifetime measurements of poly(NIPAM-co-DBDAA) in aqueous solution. The origin of fluorescence response of poly(NIPAM-co-DBDAA) to change in temperature was investigated in terms of local environment around the fluorescent DBD-AA units. A model fluorophore DBD-IA (Figure 1b) was synthesized, and its fluorescence lifetime was measured in n-hexane, 1,4-dioxane, ethyl acetate, ethanol, 2,2,2-trifluoroethanol, methanol, acetonitrile, DMSO, water, methanol-d1 (CH3OD), and deuterium oxide to elucidate influences of polarity and hydrogen bonding ability of the solvents on the photophysical properties. To know the hydrogen-bonding sites in the DBD-AA unit with solvent molecules, fluorescence lifetimes of two other reference compounds DBThD-IA (Figure 1c) and BD-DA (Figure 1d) were measured in water and deuterium oxide. These fluorescence lifetime measurements for DBD-IA, DBThD-IA, and BD-DA clarified the usefulness of the DBD-AA units for a highly sensitive fluorescent polymeric thermometer with temperaturedependent fluorescence lifetime. 2. Experimental Section Materials and Apparatus. BD-DA,19 N,N-dimethyl-7-[methyl{2-(methylamino)ethyl}amino]-2,1,3-benzoxadiazole-4-sulfonamide (DBD-NMe(CH2)2NHMe),18 and poly(NIPAM-coDBD-AA) (Mn ) 36 400, Mw/Mn ) 3.08)18 were obtained as previously reported. Isobutyric anhydride was purchased from TCI. Acetonitrile, 1,4-dioxane, DMSO, ethyl acetate, n-hexane, and methanol were of spectrophotometric grade (Dojindo). Ethanol (dehydrated, 99.5%) was purchased from Wako Pure Chemicals. 2,2,2-Trifluoroethanol was obtained from ACROS. Methanol-d1 (CH3OD) (99.5 atom % D) and deuterium oxide (99.9 atom % D) were purchased from Sigma-Aldrich. Water was purified using Milli-Q reagent systems, MILLI-Q-Labo, and Direct-Q UV (Millipore). All other reagents were of guaranteed reagent grade and used without further purification. Proton nuclear magnetic resonance (1H NMR) spectra were obtained using a Bruker AVANCE 400 spectrometer (400
Gota et al. MHz). The J values are given in hertz. Mass spectra with electrospray ionization (ESI) system were measured on a Bruker micrOTOF-05 spectrometer. Melting points were measured using a Yanagimoto Micro Melting Point Apparatus and are uncorrected. Synthesis of DBD-IA. DBD-NMe(CH2)2NHMe (34.0 mg, 109 µmol) was dissolved in acetonitrile (3 mL). After the addition of triethylamine (18.2 µL, 130 µmol) and isobutyric anhydride (27.1 µL, 163 µmol) at 0 °C, the mixture was stirred at room temperature for 40 min. Then Na2CO3 (0.5 g) was added into the reaction mixture. The resultant was filtered and evaporated to dryness under reduced pressure. The residue was chromatographed on silica gel with chloroform-methanol (20: 1) as an eluent to afford DBD-IA (36.8 mg, 88%) as an orange powder: mp, 138.5-139 °C; 1H NMR (CDCl3, δ) 7.87 (1H, d, J 8.3), 6.11 (1H, d, J 8.3), 4.27 (2H, t, J 6.6), 3.67 (2H, t, J 6.6), 3.33, 3.28 (3H, s) 3.06, 3.04 (3H, s), 2.90, 2.86 (6H, s), 2.66 (1H, q), 1.11, 0.95 (6H, d, J 6.7). Anal. Calcd for C16H25N5O4S: C, 50.11; H, 6.57; N, 18.26. found: C, 50.30; H, 6.71; N, 17.63. HR-ESI-MS m/z: Calcd for C16H26N5O4S+ ([M+H]+) 384.1705, found 384.1690. Synthesis of DBThD-IA. DBThD-NMe(CH2)2NHMe (80.0 mg, 243 µmol; see Supporting Information for the preparation) was dissolved in dichloromethane (7 mL). After the addition of triethylamine (40.6 µL, 291 µmol) and isobutyric anhydride (60.7 µL, 364 µmol), the mixture was stirred at room temperature for 15 min. Then Na2CO3 (1 g) was added into the reaction mixture, and after filtration, the resultant was poured into dichloromethane (100 mL). The organic layer was washed with 0.1 M NaOH aqueous solution (50 mL × 2) and water (50 mL × 2). The organic layer was dried over Na2SO4 and was evaporated to dryness under reduced pressure. The residue was chromatographed on silica gel with dichloromethane-methanol (200:1 to 150:1) as an eluent to afford DBThD-IA (77.0 mg, 79%) as an orange oil: 1H NMR (CDCl3, δ) 8.12, 8.08 (1H, d, J 8.4), 6.45, 6.44 (1H, d, J 8.4), 4.38, 4.31 (2H, t, J 6.8), 3.78, 3.73 (2H, t, J 6.8), 3.33, 3.25 (3H, s), 3.05 (3H, s), 2.90, 2.87 (6H, s), 2.68 (1H, q) 1.11, 0.99 (6H, d, J 6.8). HR-ESI-MS m/z: Calcd for C16H26N5O3S2+ ([M+H]+) 400.1477, found 400.1487. Photophysical Study of Poly(NIPAM-co-DBD-AA) in Aqueous Solution. Fluorescence spectra of poly(NIPAM-coDBD-AA) were obtained using a JASCO FP-6500 spectrofluorometer with a Hamamatsu R-7029 optional photomultiplier tube (operative range, 200-850 nm) and were corrected by using a JASCO ESC-333 substandard light source. The temperature was controlled by a JASCO ETC-273T temperature controller. The copolymers were excited at 456 nm, which is the maximum excitation wavelength of the DBD-AA units. The samples were equilibrated at each temperature for at least 2 min. The temperature of the sample solution was monitored with a platinum resistance thermometer. Fluorescence lifetime of poly(NIPAM-co-DBD-AA) was measured with a time-correlated single-photon counting fluorimeter (Edinburgh Analytical Instruments, FL-900CDT). A nanosecond pulsed discharge lamp (pulse width, 1 ns; repetition rate, 40 kHz) filled with hydrogen gas was used as an excitation light source. The sample was excited at 275 nm. Its temperature was controlled by a water bath and was measured by a thermistor before and after each measurement. The fluorescence decay curve obtained was best fitted by a double exponential function
I(t) ) B1 exp(-t/τ1) + B2 exp(-t/τ2)
(1)
Fluorescence Lifetime of NIPAM
J. Phys. Chem. B, Vol. 112, No. 10, 2008 2831
Figure 2. (a) Fluorescence decay curves (Ex, 275 nm; Em, 560 nm) and (b) fluorescence spectra (Ex, 456 nm) of poly(NIPAM-co-DBD-AA) in aqueous solution (0.02 w/v%). In the region of dotted lines, scatter due to excitation light overlapped with the fluorescence spectra.
Fractional contributions (P1 and P2 for τ1 and τ2, respectively) and average fluorescence lifetime 〈τf〉 was calculated by using eqs 2-4
Temperature resolution δT (°C) between T and T + ∆T (°C) was evaluated from 〈τf〉 values by using eq 5, which is similar to the fluorescence intensity as defined in the previous study18
()
P1 )
B1τ1 × 100 B1τ1 + B2τ2
(2)
P2 )
B2τ2 × 100 B1τ1 + B2τ2
(3)
B1τ12 + B2τ22 B1τ1 + B2τ2
(4)
x12 {[τ (T + ∆T) - τ(T + ∆T)] + [τ (T) - τ(T)] } (5)
The relative standard deviation for the fluorescence lifetime measurements was estimated to be ca. 2.6% from ten repetitive measurements using poly(NIPAM-co-DBD-AA) in water at 40 °C.
where ∂T/∂τ and δτ represent the inverse of the slope in an average fluorescence lifetime-temperature diagram and the standard deviation of average fluorescence lifetime, respectively, τH (T) and τC (T) represent average fluorescence lifetime at
〈τf〉 )
δT )
∂T δτ ) ∂τ
(
(T + ∆T) - T
τ(T + ∆T) - τ(T)
)
2
H
2
H
2832 J. Phys. Chem. B, Vol. 112, No. 10, 2008
Gota et al.
T °C during heating and cooling, respectively, and τ(T) is equal to (τH (T) + τC (T))/2. Photophysical Study of DBD-IA, DBThD-IA, and BD-DA in Various Solvents. Absorption and fluorescence spectra were measured in n-hexane, 1,4-dioxane, ethyl acetate, ethanol, 2,2,2trifluoroethanol, methanol, methanol-d1, acetonitrile, DMSO, water, deuterium oxide, and the mixtures of 1,4-dioxane and water (5:5, 3:7, and 1:9, v/v) (for results, see Supporting Information). UV-visible absorption spectra (5-30 µM) were measured using a JASCO V-550 UV/vis spectrophotometer. Fluorescence spectra (1-10 µM) were obtained with the same equipment as described above. The fluorescence quantum yields (Φf) were determined by the following equation
Φf,S ) Φf,RFSARnS2/FRASnR2
(6)
where F is the area under the fluorescence spectrum with excitation at 458 nm, A is the absorbance at 458 nm, n is the refractive index of the solvent, and the subscripts R and S represent reference and sample, respectively. 4-Methylamino7-nitoro-2,1,3-benzoxadiazole in acetonitrile (Φf ) 0.38 with excitation at 458 nm20) was used as a reference. Fluorescence lifetimes were measured in the same solvents used for the measurements of absorption and fluorescence spectra. FL-900CDT (Edinburgh Analytical Instruments) was used for nanosecond lifetime measurements. Picosecond timeresolved fluorescence measurements were carried out by using a femtosecond laser system that was based on a mode-locked Ti:sapphire laser (Spectra-Physics, Tsunami; center wavelength, 800 nm; pulse width, ∼70 fs; repetition rate, 82 MHz) pumped by a CW green laser (Spectra-Physics, Millenia V; 532 nm, 4.5 W). The generation of the second harmonic (400 nm; pulse width, ∼200 fs) was performed in an LBO crystal. The third harmonic (266 nm; pulse width, ∼250 fs) was generated by a sum frequency mixing of the fundamental and the second harmonic of the Tsunami laser system. The repetition frequency of the excitation pulse was reduced to 4 MHz by using a pulse picker (Spectra-Physics Model 3980). The second harmonic (400 nm) in the output beam was used as a trigger pulse. The emission light was detected by a microchannel plate photomultiplier (Hamamatsu R3809U-51) after passing through a monochromator (Oriel Model 77250). The instrumental response function had a half-width of 20-25 ps. The fluorescence time profiles were analyzed by iterative reconvolution with the response function. For cases showing a multiexponential decay curve, average fluorescence lifetime 〈τf〉 was calculated with foregoing eq 4. 3. Results 3.1. Fluorescence Lifetime of a Fluorescent Polymeric Thermometer, Poly(NIPAM-co-DBD-AA), at Various Temperatures. First, fluorescence lifetime of poly(NIPAM-co-DBDAA) in aqueous solution was measured at various temperatures in the range from 25 to 40 °C, in which temperature-dependent fluorescence lifetime could be expected since a structural change of polyNIPAM units from an extended form to a globular form occurs around 32 °C.21 Figure 2 shows representative fluorescence decay curves of poly(NIPAM-co-DBD-AA) as well as its fluorescence spectra, and Table 1 summarizes the results of the fluorescence lifetime measurements. It should be noted that the fluorescence decay curves at the temperatures between 25 and 40 °C were well fitted with double-exponential functions but not with single-exponential ones. Figure 3 indicates a
TABLE 1: Fluorescence Lifetimes of Poly(NIPAM-co-DBD-AA) at Various Temperatures (0.02 w/v% in Water; Ex, 275 nm; Em, 560 nm) temperature/°C
τ1/ns
P1a/%
τ2/ns
P2b/%
〈τf〉c/ns
25 27 29 30 31 31.5 32 33 35 40
1.90 2.03 1.91 2.08 2.19 2.80 3.00 3.02 3.91 7.09
53 55 49 53 51 38 16 7.6 6.4 16
6.22 6.52 6.28 6.61 7.79 13.1 14.4 14.6 14.7 15.4
47 45 51 47 49 62 84 92 94 84
3.92 4.06 4.10 4.22 4.96 9.18 12.6 13.7 14.1 14.1
a Composition of τ1. b Composition of τ2. c Average fluorescence lifetime calculated by eq 4.
lifetime component (τ1 and τ2, τ1
