J. Phys. Chem. B 2009, 113, 13917–13925
13917
Nanosecond to Submillisecond Dynamics in Dye-Labeled Single-Stranded DNA, As Revealed by Ensemble Measurements and Photon Statistics at Single-Molecule Level Takahiro Kaji,† Syoji Ito,†,‡ Shigenori Iwai,† and Hiroshi Miyasaka*,†,§ DiVision of Frontier Materials Science and DiVision of Chemistry, Graduate School of Engineering Science and Center for Quantum Materials Science under Extreme Conditions, Osaka UniVersity, Machikaneyama1-3, Toyonaka, Osaka 560-8531, Japan; PRESTO, Japan Science and Technology Agency, Tokyo, Japan; and CREST, Japan Science and Technology Agency, Tokyo, Japan ReceiVed: December 17, 2008; ReVised Manuscript ReceiVed: August 30, 2009
Single-molecule and ensemble time-resolved fluorescence measurements were applied for the investigation of the conformational dynamics of single-stranded DNA, ssDNA, connected with a fluorescein dye by a C6 linker, where the motions both of DNA and the C6 linker affect the geometry of the system. From the ensemble measurement of the fluorescence quenching via photoinduced electron transfer with a guanine base in the DNA sequence, three main conformations were found in aqueous solution: a conformation unaffected by the guanine base in the excited state lifetime of fluorescein, a conformation in which the fluorescence is dynamically quenched in the excited-state lifetime, and a conformation leading to rapid quenching via nonfluorescent complex. The analysis by using the parameters acquired from the ensemble measurements for interphoton time distribution histograms and FCS autocorrelations by the single-molecule measurement revealed that interconversion in these three conformations took place with two characteristic time constants of several hundreds of nanoseconds and tens of microseconds. The advantage of the combination use of the ensemble measurements with the single-molecule detections for rather complex dynamic motions is discussed by integrating the experimental results with those obtained by molecular dynamics simulation. Introduction Conformational motions of large biomolecules such as nucleic acids and proteins take an important role in their functionalities.1 In order to directly detect these motions, dye-labeled systems have been widely utilized. Information on the molecular motions has been deduced from the measurement and analysis of the fluorescent properties, such as intensity, lifetime, and wavelength, of the dye molecule connected covalently or noncovalently to the appropriate position of the biomacromolecule.2 In general, photoinduced electron transfer (PET) and Fo¨rster resonance energy transfer (FRET) are important channels resulting in the change of the fluorescent properties.3 The observation of fluorescence quenching induced by PET indicates the existence of the conformation where the fluorescent dye locates in the vicinity of the quencher. The decrease or increase in the fluorescence intensities of the donor and acceptor for the excitation energy corresponds to the geometrical change that enables the transfer of the energy to occur. Hence, the detection of the dynamic behaviors of the fluorescence provides the information on the time scale of these conformational motions. In the case that the conformational change affecting the fluorescence occurs within the fluorescent lifetime of the dye, we can directly deduce the time constant of the conformational motion from the difference between the fluorescent lifetime in the presence of the quencher and that without the quencher, on the basis of the ensemble measurement.4-8 The lifetime of a typical dye, however, stays in the range of a few ns to several tens of nanoseconds, and it is generally difficult to monitor the * Corresponding author. Phone: +81-6-6850-6241. Fax: +81-6-68506244. E-mail:
[email protected]. † Osaka University. ‡ PRESTO. § CREST.
conformational motions longer than nanosecond time scale by the ensemble time-resolved measurements. An important aspect of the macromolecules is that they represent complex molecular dynamics over a broad range of time scales, including the fast local motions and the slow changes in large scales. Hence, it is indispensable to scan wide time region to comprehensively elucidate these hierarchical motions. Although the triplet quenching9-12 could provide the information on the conformational motions in longer time scale, it is not always applicable to the various macromolecular systems because of the weak emission of phosphorescence. To overcome these restrictions and to obtain the information in a wide time region, a method was proposed on the basis of the detection of the fluorescence in the single molecular level in dilute solution by using a confocal optical microscope.13-19 In this method, the fluctuation of the intensity of the fluorescence from single molecules in the confocal volume is monitored by the detector with rather high time resolution, typically < nanoseconds. The detected signals are usually converted into statistical functions that are suitable for further analysis to obtain the quantitative information on the conformational dynamics of the molecules, of which information is usually difficult to acquire by means of the ensemble time-resolved measurement. Conventionally, two types of statistical functions were employed for the analysis. Time interval for the consecutive photons of fluorescence provides the interphoton arrival time histogram, which has been used as an autocorrelation function in the time region of subnanoseconds to microseconds.15,16,19,20 On the other hand, a fluorescence intensity autocorrelation function that is based on the fluorescence correlation spectroscopy (FCS) has been used for the time range from submicroseconds to milliseconds. In recent years, the time range of the FCS has been extended to shorter region of nanoseconds with
10.1021/jp811122a CCC: $40.75 2009 American Chemical Society Published on Web 09/25/2009
13918
J. Phys. Chem. B, Vol. 113, No. 42, 2009
progresses in experimental apparatus.21 A full autocorrelation function from subnanoseconds to milliseconds or combination of the interphoton arrival time histogram and the fluorescence intensity autocorrelation function have been employed for the investigation of macromolecular dynamics spanning over a broad range of time scales.16,19,21 In particular, the analysis of autocorrelation and cross correlation between donor and acceptor fluorescence intensities has provided precise information on the dynamics of the dye-labeled complex systems.16,19 Although the combined use of the autocorrelation with the cross correlation is an excellent and powerful method for the investigation of conformational motions in various systems,14,16,19 it usually requires the specimen with the excitation energy donor and acceptor pair. Hence, this excellent method is not straightforwardly applicable to molecular systems where the simple fluorescence quenching such as PET takes place. Compared to the excitation energy transfer, the electron transfer reaction22,23 is more severely dependent on the distance between the fluorophore and the quencher. The dependence of the electron transfer rate constant22 on the pair distance, r, is represented by k(r) ) k(0) exp(-r/β) and the typical value of the dumping factor,22 β, is ca. 1 Å. On the other hand, the distance dependence24 for the Fo¨rster excitation energy transfer is represented by k(r) ) k(r0)r06/r6 and the typical value of r0 is 10-30 Å. In other words, the rate constant of electron transfer at 20 Å is ca. 10-9 of k(0) at the close contact of the donor and acceptor, while we can find many donor-acceptor systems of the excitation energy transfer with the rate constant of ca.109/s even at 20 Å distance. Hence, the investigation by using PET may potentially provide more precise information on geometries and their dynamics. In the present work, with the aim to precisely elucidate the conformational dynamics by observing the simple quenching system of PET, we have investigated the fluorescence dynamics of the single-stranded DNA, ssDNA, connected with a fluorescein dye by a C6 linker. The linker-attached fluorophores have been widely used for the detection of DNA/RNA sequences,25,26 the geometry27,28 and conformational motion28-32 of DNA molecules, and so forth.33-38 Although the detailed information on the time scale of conformational dynamics is quite important for the investigation in these DNAs, rather large flexibility of C6 linker makes it rather difficult to clearly elucidate the local conformational motions of the linker part as well as a whole conformational motion. To acquire the detailed information on the dynamics, we have applied the analysis of the autocorrelation function and interphoton arrival time histogram from nanoseconds to submilliseconds, by using various values such as the number and fraction of conformations and kinetic parameters obtained by dynamic and steady-state ensemble measurements of the fluorescence quenching via the electron transfer between the dye and a guanine base in the ssDNA. Two distinctive time scales of conformational dynamics, that is, the dynamics of several hundreds of nanoseconds and that of tens of microseconds, were characterized from the analysis, in which three conformations with different fluorophore-quencher distances were involved. Molecular dynamics simulation suggested the existence of not only a conformation with a long distance between the dye and the base but also a conformation with a short distance between them, which could lead to the formation of a nonfluorescent complex. In the following, we will discuss the advantage of the combination use of the ensemble measurements with the single-molecule detections for rather complex dynamic motions by integrating the experimental results with those obtained by molecular dynamics simulation.
Kaji et al.
Figure 1. Sequences of single-stranded DNAs (G0-G4). The thymine base indicated as T‡ (fluorescein-dT) possesses fluorescein coupled via C6 linker.
Experimental Section Single-Stranded DNAs. Sequences of 11-mer oligonucleotides used in the present work are shown in Figure 1. The thymine base, indicated as T‡, has a fluorescein dye coupled via C6 linker. G1-G4 has a guanine base at different positions, while G0 has no guanine base in the sequence. The dye-labeled thymine base (fluorescein-dT phosphoramidite, Glen Research) was incorporated by DNA synthesizer (394 DNA/RNA synthesizer, Applied Biosystems). After the deprotection, the oligonucleotides were purified by reversed-phase high-performance liquid chromatography (m-Bondasphere C18 15 µm 300 A, Waters). All the measurements for ssDNAs were performed in 30 mM phosphate buffer solution at pH 7.0 containing 100 mM NaCl. The concentration of G0-G4 was adjusted to 1.67 × 10-6 M (absorbance ) 0.1 at absorption maximum of fluorescein) for steady-state fluorescence measurement and timecorrelated single-photon counting. On the other hand, diluted solution of 1.0 × 10-9 M was used for the measurements in the single-molecule level. Steady-State Fluorescence Measurement and Time-Correlated Single-Photon Counting. Steady-state fluorescence and absorption spectra were respectively measured by Hitachi 850 fluorescence spectrophotometer and Hitachi U-3500 spectrophotometer. For time-resolved fluorescence measurement, a time-correlated single-photon-counting (TCSPC) method was employed using a mode-locked femtosecond Ti:sapphire laser (Tsunami, Spectra Physics) with 8 MHz repetition for the excitation source. SHG at 488 nm was employed for the excitation. The fluorescence at the magic angle was detected by a microchannel plate photomultiplier tube (R3809U-50, Hamamatsu) after passing through a monochromator to monitor
Conformational Dynamics of ssDNA the fluorescence at 520 nm. A full width at half-maximum of the instrumental response function was 34 ps. For the evaluation of the analyzed result, we used a reduced χ2 value.39,40 Experimental Setup for Fluorescence Correlation Spectroscopy and Photon Statistics. The fluorescence in the singlemolecule level was detected by using a confocal microscope coupled to two single-photon-counting detectors with subnanosecond time resolution. For the excitation, an output of circularly polarized cw 488 nm Ar+ laser (LGK7872M, LASOS Lasertechnik GmbH) was led into an inverted microscope (IX71, Olympus) with a dichroic mirror and was focused by an objective lens (UPlan FL N 100×, Olympus). The fluorescence light was collected by the same objective lens and transmitted through the dichroic mirror and a band-pass filter (FF01-536/ 40-25, Semrock). A 50 µm pinhole (P50S, Thorlabs) was introduced to ensure the detection from the confocal volume. Then the fluorescence light was separated by a 50:50 nonpolarizing beam splitter and was guided into two avalanche photodiodes (SPCM-AQR-14, Perkin-Elmer) for the detection. These two detectors were connected to a time-correlated singlephoton-counting module (Picoharp 300, Picoquant GmbH) that recorded photon arrival times from the two avalanche photodiodes with a time interval of 4 ps for the whole recording time. After sending these data into PC, an interphoton arrival time histogram in the time range of 0.5 ms with a bin time of 1 ns was obtained by homemade software. Because the background signal of the buffer solution stayed in the level of the dark noise of the detector ( G3 > G2 > G1). This result may be due to the difference in the accessibility of the dye to the guanine in the ssDNAs, which is influenced by the flexibility of the C6 linker and the ssDNAs. The time constants for the interconversion between forms F and D were in the time scale of several hundreds of nanoseconds. These values were similar to the time constants of conformational change of methylene chains or small peptide chains with similar length obtained by ensemble measurements and a singlemolecule measurement in the earlier reports.4-12 Hence, it is strongly suggested that the present interconversion between F and D forms is mainly ascribable to the local motion of C6 linker. On the other hand, the slow conformational dynamics (tens of microseconds) between D and C forms is attributable to the whole conformational motion of the DNA. To obtain more detailed information on these conformations, we employed the molecular dynamics (MD) simulation. In this calculation, we used the model systems of 5-mer oligonucleotide
TABLE 3: Parameters Obtained from Analysis of Interphoton Arrival Time Histograms G1 G2 G3 G4
(18)
kFfD/s-1 (1/kFfD/ns)
kDfF/s-1 (1/kDfF/ns)
kDfC/s-1 (1/kDfC/µs)
kCfD/s-1 (1/kCfD/µs)
4.0 × 105 (2500) 8.0 × 105 (1300) 1.6 × 106 (630) 4.0 × 106 (250)
2.8 × 106 (350) 6.3 × 106 (160) 9.9 × 106 (100) 1.8 × 107 (54)
1.0 × 105 (10) 2.0 × 104 (50) 8.0 × 104 (13) 1.6 × 105 (6.3)
1.4 × 105 (7.0) 7.5 × 104 (13) 8.8 × 104 (11) 1.2 × 105 (8.3)
13924
J. Phys. Chem. B, Vol. 113, No. 42, 2009
Kaji et al.
Figure 9. A trajectory of distance between a fluorescein (F) and a guanine base (G) of G4, obtained by molecular dynamics simulation at 300 K (a). The representative structures of G4 that are indicated by arrows in the trajectory (b-d). The sampling time interval was 50 ps.
Figure 8. Interphoton arrival time histograms (colored circles) and calculated results of φ(τ) (black lines) for G1-G4. Gray and red lines are the contribution of GF(τ) and that of GD(τ) (see text). The accumulation time was 10 h.
molecules containing four cytosine bases, a guanine base, and a fluorescein-dT (for example, 5′- CCC GT‡-3′ for G1) to shorten the simulation time. A trajectory of the distance between the centers of the fluorescein and the guanine base of G4 at 300 K and the representative structures in the trajectory are shown in Figure 9. The initial conformation at the time origin was set to be a rather extended structure as shown in Figure 9b. The temporal trajectory shows that the short distance conformation and the long distance between the fluorescein and the guanine base were kept for more than a few tens of nanoseconds. A similar result was obtained for the other ssDNAs (see Supporting Information). Because of the short time window of the simulation time in addition to the effect of elimination of 3′ part of the molecule, we cannot conclude whether these conformations with close distance are ascribable to form D or C. However, the duration of these forms for more than the lifetime of the dye (>4 ns) in the simulation result is consistent with the experimental result that the forms interconvert with the time constant of more than the lifetime of the dye. Figure 10 shows the distributions of the distance between the fluorescein and the guanine base for G1-G4, where the contribution from the conformations with the short distance is large in G3 and G4. Although it is necessary to extend the time window for the elucidation of the precise conformational dynamics, the present results of MD could support the long staying time of the form C leading to the static and/or rapid quenching at room temperature.
Figure 10. Distributions of distance between a fluorescein and a guanine base of G1-G4, obtained by molecular dynamics simulation at 300 K. The sampling time interval and the simulation time were 50 ps and 250 ns, respectively.
Concluding Remarks In the present study, we have applied the analysis of the autocorrelation function and interphoton arrival time histogram in the nanosecond to submillisecond time region using parameters obtained from the ensemble measurements of the fluorescence quenching via electron transfer between the dye and the guanine base to investigate the conformational dynamics of fluorescein-labeled single-stranded DNAs. Although rather complicated model with the seven-state and three conformations was employed on the basis of the ensemble measurements, the time constants of the interconversion among these three conformations well reproduced the experimental results. That is, those of the interconversion between the form F (nonquenching conformation) and the form D (dynamically quenching conformation) were in the order of several hundreds of nanoseconds, and they decreased with an increase in the distance between the dye and the base in the sequence. The difference in the time constants may be ascribable to the accessibility of
Conformational Dynamics of ssDNA the dye to the base. On the other hand, the time constants of the interconversion between the form D and the form C (rapid or static quenching conformation) were estimated to be in the order of tens of microseconds, and these time constants were confirmed to be consistent with the time constants by the FCS. MD simulation suggested that these forms existed for the duration of more than the fluorescence lifetime and the fraction of forms D or C was large in G3 and G4. Our approach on the basis of the results of the single-molecule and ensemble measurements is useful for the investigation of dynamics of macromolecules in the time range of nanosecond to submillisecond even in the case that a simple fluorescence quenching takes place with complicated kinetics. Supporting Information Available: Full description of intermolecular quenching experiment and estimation of energy gap for the charge separation, details of the estimation of absorption cross section, fluorescence intensity autocorrelation functions of the other samples, and the other results of molecular dynamics simulations at 300 and 600 K. This material is available free of charge via the Internet at http://pubs.acs.org. Acknowledgment. This work was partly supported by Grantin-Aids for Scientific Research (19050006 and 20350009) from the Ministry of Education, Culture, Sports, Science, and Technology (MEXT) of the Japanese Government, and the 21st Century COE Program “Core Research and Advanced Education Center for Materials Science and Nano Engineering” from the Japan Society for the Promotion of Science (JSPS). We appreciate Prof. N. Tamai at Kwansei Gakuin University for his valuable discussion and advice for the analysis of the results of the time-correlated single-photon counting. References and Notes (1) McCammon, J. A.; Harvey, S. C. Dynamics of proteins and nucleic acids; Cambridge University Press: Cambridge, UK, 1987. (2) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 3rd ed.; Springer: New York, 2006. (3) Michalet, X.; Weiss, S; Ja¨ger, M. Chem. ReV. 2006, 106, 1785. (4) Hudgins, R. R.; Huang, F.; Gramlich, G.; Nau, W. M. J. Am. Chem. Soc. 2002, 124, 556. (5) Hirayama, F. J. Chem. Phys. 1965, 42, 3163. (6) Okada, T.; Fujita, T.; Kubota, M.; Masaki, S.; Mataga, N. Chem. Phys. Lett. 1972, 14, 563. (7) Migita, M.; Okada, T.; Mataga, N.; Sakata, Y.; Misumi, S.; Nakashima, N.; Yoshihara, K. Bull. Chem. Jpn. 1981, 54, 3304. (8) Okada, T.; Migita, M.; Mataga, N.; Sakata, Y.; Misumi, S. J. Am. Chem. Soc. 1981, 103, 4715. (9) Sinclair, A. M.; Winnik, M. A. J. Am. Chem. Soc. 1985, 107, 5798. (10) Miyasaka, H.; Kiri, M.; Morita, K.; Mataga, N.; Tanimoto, Y. Bull. Chem. Jpn. 1995, 68, 1569. (11) Wagner, P. J.; Kla´n, P. J. Am. Chem. Soc. 1999, 121, 9626. (12) Lapidus, L. J.; Eaton, W. A.; Hofrichter, J. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 7220. (13) Hanbury Brown, R.; Twiss, R. Q. Nature 1956, 177, 27. (14) Berglund, A. J.; Doherty, A. C.; Mabuchi, H. Phys. ReV. Lett. 2002, 89, 068101.
J. Phys. Chem. B, Vol. 113, No. 42, 2009 13925 (15) Neuweiler, H.; Schulz, A.; Bo¨hmer, M.; Enderlein, J.; Sauer, M. J. Am. Chem. Soc. 2003, 125, 5324. (16) Nettels, D.; Gopich, I. V.; Hoffmann, A.; Schuler, B. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 2655. (17) Doose, S.; Neuweiler, H.; Barsch, H.; Sauer, M. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 17400. (18) Kim, J.; Doose, S.; Neuweiler, H.; Sauer, M. Nucleic Acids Res. 2006, 34, 2516. (19) Nettels, D.; Hoffmann, A.; Schuler, B. J. Phys. Chem. B 2008, 112, 6137. (20) Kimble, H. J.; Dagenais, M.; Mandel, L. Phys. ReV. Lett. 1977, 39, 691. (21) Felekyan, S.; Ku¨hnemuth, R.; Kudryavtsev, V.; Sandhagen, C.; Becker, W.; Seidel, C. A. M. ReV. Sci. Instrum. 2005, 76, 083104. (22) Marcus, R. A.; Sutin, N. Biochim. Biophys. Acta 1985, 811, 265. (23) Mataga, N.; Miyasaka, H. AdV. Chem. Phys. 1999, 107, 431. (24) (a) Fo¨rster, T. Ann. Phys. 1948, 2, 55. (b) Fo¨rster, T. In Modern Quantum Chemistry; Sinanoglu, O., Ed.; Academic Press: New York, 1965; pp 93. (25) Heyduk, T.; Heyduk, E. Nat. Biotechnol. 2002, 20, 171. (26) Silverman, A. P.; Kool, E. T. Nucleic Acids Res. 2005, 33, 4978. (27) Norman, D. G.; Grainger, R. J.; Uhrı´n, D.; Lilley, D. M. J. Biochemistry 2000, 39, 6317. (28) Neubaur, H.; Gaiko, N.; Berger, S.; Schaffer, J.; Eggeling, C.; Tuma, J.; Verdier, L.; Seidel, C. A. M.; Griesinger, C.; Volkmer, A. J. Am. Chem. Soc. 2007, 129, 12746. (29) Sauer, M.; Drexhage, K. H.; Lieberwirth, U.; Mu¨ller, R.; Nord, S.; Zander, S. Chem. Phys. Lett. 1998, 284, 153. (30) Eggeling, C.; Fries, J. R.; Brand, L.; Gu¨nther, R.; Seidel, C. A. M. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 1556. (31) Jung, J.; Ihly, R.; Scott, E.; Yu, M.; Orden, A. V. J. Phys. Chem. B 2008, 112, 127. (32) Heinlein, T.; Knemeyer, J.-P.; Piestert, O.; Sauer, M. J. Phys. Chem. B 2003, 107, 7957. (33) Sandin, P.; Wilhelmsson, L. M.; Lincoln, P.; Powers, V. E. C.; Brown, T.; Albinsson, B. Nucleic Acids Res. 2005, 33, 5019. (34) Nazarenko, I.; Pires, R.; Lowe, B.; Obaidy, M.; Rashtchian, A. Nucleic Acids Res. 2002, 30, 2089. (35) Va´mosi, G.; Gohlke, C.; Clegg, R. M. Biophys. J. 1996, 71, 972. (36) Famulok, M.; Hartig, J. S.; Mayer, G. Chem. ReV. 2007, 107, 3715. (37) Sparano, B. A.; Koide, K. J. Am. Chem. Soc. 2007, 129, 4785. (38) Unruh, J. R.; Gokulrangan, G.; Wilson, G. S.; Johnson, C. K. Photochem. Photobiol. 2005, 81, 682. (39) O’Connor, D. V.; Phillips, D. Time-Correlated Single-Photon Counting; Academic Press: London, 1984. (40) Boens, N.; Qin, W.; Basariæ, N.; Hofkens, J.; Ameloot, M.; Pouget, J.; Lefe`vre, J.-P.; Valeur, B.; Gratton, E.; vandeVen, M.; Silva, N. D.; Engelborghs, Y.; Willaert, K.; Sillen, A.; Rumbles, G.; Phillips, D.; Visser, A. J. W. G.; Van Hoek, A.; Lakowicz, J. R.; Malak, H.; Gryczynski, I.; Szabo, A. G.; Krajcarski, D. T.; Tamai, N.; Miura, A. Anal. Chem. 2007, 79, 2137. (41) Wang, J.; Cieplak, P.; Kollman, P. A. J. Comput. Chem. 2000, 21, 1049. (42) Wang, J.; Wolf, R. M.; Caldwell, J. W.; Kollman, P. A.; Case, D. A. J. Comput. Chem. 2004, 25, 1157. (43) Qui, D.; Shenkin, P. S.; Hollinger, F. P.; Still, W. C. J. Phys. Chem. A 1997, 101, 3005. (44) Seidel, C. A. M.; Schulz, A.; Sauer, M. H. M. J. Phys. Chem. 1996, 100, 5541. (45) Torimura, M.; Kurata, S.; Yamada, K.; Yokomaku, Y.; Kamagata, T.; Kanagawa, T.; Kurane, R. Anal. Sci. 2001, 17, 155. ¨ . J. Fluoresc. 1994, 4, 255. (46) Widengren, J.; Rigler, R.; Mets, U ¨ .; Rigler, R. J. Phys. Chem. 1995, 99, 13368. (47) Widengren, J.; Mets, U (48) Widengren, J.; Dapprich, J.; Rigler, R. Chem. Phys. 1997, 216, 417. (49) Krichevsky, O.; Bonnet, G. Rep. Prog. Phys. 2002, 65, 251. (50) Gopich, I. V.; Szabo, A. J. Chem. Phys. 2006, 124, 154712.
JP811122A
