- Journal List
- HHS Author Manuscripts
- PMC6880806
As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsem*nt of, or agreement with, the contents by NLM or the National Institutes of Health.
Learn more: PMC Disclaimer | PMC Copyright Notice
J Biochem Mol Biol. Author manuscript; available in PMC 2019 Nov 27.
Published in final edited form as:
J Biochem Mol Biol. 2002 Jul 31; 35(4): 389–394.
PMCID: PMC6880806
NIHMSID: NIHMS1059738
PMID: 12296998
Jung Sook Kang,* Omoefe O. Abugo,† and Joseph R. Lakowicz†
Author information Copyright and License information PMC Disclaimer
Abstract
[Ru(bpy)2(dppz)]2+ (bpy = 2,2’-bipyridine,dppz = dipyrido-[3,2-a:2’,3’-c]phenazine) (RuBD), a long-lifetimemetal-ligand complex, displays favorable photophysical properties. These includelong lifetime, polarized emission, but no significant fluorescence from thecomplex that is not bound to DNA. To show the usefulness of this luminophore(RuBD) for probing the bending and torsional dynamics of nucleic acids, itsintensity and anisotropy decays when intercalated into supercoiled and relaxedpTZ18U plasmids were examined using frequency-domain fluorometry with a bluelight-emitting diode (LED) as the modulated light source. The mean lifetimes forthe supercoiled plasmids (< τ > = 148 ns) were somewhatshorter than those for the relaxed plasmids (< τ > = 160ns). This suggests that the relaxed plasmids were shielded more efficiently fromwater. The anisotropy decay data also showed somewhat shorter slow rotationalcorrelation times for supercoiled plasmids (288 ns) than for the relaxedplasmids (355 ns). The presence of two rotational correlation times suggeststhat RuBD reveals both the bending and torsional motions of the plasmids. Theseresults indicate that RuBD can be useful for studying both the bending andtorsional dynamics of nucleic acids.
Keywords: Bending and torsional dynamics, Light-emitting diode, Long-lifetime metal-ligand complex, Supercoiled and relaxed pTZ18U
Introduction
The structure and dynamics of DNA are receiving more and more attention. Theyhave been studied by various techniques including time-resolved fluorescenceanisotropy (Schurr et al.,1992; Kielkopf et al.,2000; Lee and Choi, 2000; Okonogi et al., 2000). Theinteraction of transition metal-ligand complexes (MLCs) with DNA is an area ofintense current interest. This is partly because of the potential of these compoundsas novel probes of DNA structure and dynamics (Lakowicz et al., 1995; Malak et al., 1997; Lakowicz et al., 2000; Rajski et al., 2000; Kang and Lakowicz, 2001; Kang etal., 2002). Long-lifetime MLCs, which display decay timesthat range from 100 ns to more than 10 μs, have only recently becomeavailable (DeGraff et al.,1994; Terpetschnig etal., 1997; Lakowicz etal., 2000). They have favorable chemical, photochemical, andphotophysical properties. They show large Stokes’ shifts, polarized emission,as well as good water solubility and high chemical and photochemical stability(Terpetschnig et al.,1997; Lakowicz et al.,2000). In addition, the long lifetimes of the MLCs allow the use of gateddetection, which can provide increased sensitivity (Haugen and Lytle, 1981).
Barton and co-workers (Friedman etal., 1990; Jenkin etal., 1992; Murphy andBarton, 1993) reported that thedipyrido[3,2-a:2’,3’-c]phenazine (dppz) complexes of ruthenium appearto be a prime candidate for a spectroscopic probe for nucleic acids because of their“molecular light switch” properties for DNA. Since the luminescentenhancement upon DNA binding is ≥104, compared to an enhancementof ∼20 for ethidium bromide (EB), there is essentially no background with thedppz complexes of ruthenium. Barton and co-workers (Jenkin et al., 1992) also showed that the2,2’-bipyridine (bpy) derivative of the ruthenium complex,[Ru(bpy)2(dppz)]2+ (RuBD), exhibited sensitivity toconformational differences in DNA because of the incomplete shielding of the dppzligand from water in the presence of bpy, in contrast to the other phenanthrolinederivative. In this laboratory, we introduced the use of fluorescence anisotropydecays of the dppz complexes of ruthenium to study the dynamics of nucleic acids,extending the measurable time scale of DNA dynamics to the submicrosecond range.Because these complexes display large fundamental anisotropies(r0), we measured theanisotropy decays of calf thymus DNA (Lakowiczet al., 1995; Malaket al., 1997), as well as supercoiled and linearpTZ18U plasmids (Kang et al.,2002) using these ruthenium complexes.
In this study, we examined the intensity and anisotropy decays of RuBDintercalated into supercoiled and relaxed pTZ18U plasmids from E.coli HB101 in order to further show the usefulness of this luminophore(RuBD) for probing the bending and torsional dynamics of DNA. We usedfrequency-domain (FD) fluorometry with a high-intensity, blue light-emitting diode(LED) as the modulated light source. With this LED, we were able to directlymodulate the excitation light up to 100 MHz without the need for an externalmodulator (like a Pockels cell) and to obtain very reliable time-resolved intensityand anisotropy decays of RuBD, which has a very low quantum yield(Q = 0.008) (Lakowiczet al., 2001).
Materials and Methods
Materials
E. coli HB101, topoisomerase I, and bovine serum albumin(BSA) were purchased from Gibco BRL (Life Technologies, Grand Island, USA);LB-medium from Bio 101, Inc. (Vista, USA); agarose and ampicillin from Sigma(St. Louis, USA); pTZ18U plasmid from Bio-Rad (Hercules, USA); and a plasmidmega kit from Qiagen Inc. (Valencia, USA). RuBD was synthesized by the methoddescribed previously (Lakowicz etal., 1995). The chemical structure of RuBD is shown inFig. 1. All of the other chemicals wereof reagent grade. Water was deionized with a MilliQ system. All of themeasurements were carried out in a Topo-buffer (5 mM Tris-HCl, 10 mMMgCl2, 50 mM KCl, 0.5 mM dithiothreitol, 0.1 mM EDTA, and 30μg/ml BSA, pH 7.4).
Fig. 1.
Chemical structure of [Ru(bpy)2(dppz)]2+(RuBD).
Absorption and steady-state fluorescence measurement
The pTZ18U plasmids were purified with a Qiagen plasmid mega kit from 500ml overnight cultures of E. coli HB101 in LB medium thatcontained ampicillin. The pTZ18U plasmids were relaxed with topoisomerase I inTopo-buffer for 2 h at 37°C. The completion of relaxing was monitored byelectrophoresis in 1% agarose gel (Fig. 2).About 5–10 mM stock solution of RuBD was prepared in dimethylformamide.The plasmid concentration was 300 μM base pairs (bps), while theconcentration of RuBD was 15 μM. The DNA and RuBD concentrations weredetermined using molar extinction coefficients of 13,300M−1cm−1 (expressed as bp) at 260 nm and13,000 M−1cm−1 at 440 nm, respectively.UV-visible absorption spectra were measured with a Hewlett-Packard 8453 diodearray spectrophotometer. Steady-state intensity and anisotropy measurements werecarried out using an Aminco Bowman series 2 luminescence spectrometer(Spectronic Instruments, Inc., Rochester, USA). RuBD was nonfluorescent inTopo-buffer without pTZ18U plasmids in the absence and presence of topoisomeraseI. This demonstrates the absence of any impurities, as well as any effect oftopoisomerase I on the fluorescence intensity of RuBD (data not shown).
Fig. 2.
Agarose gel electrophoresis pattern of supercoiled and relaxed pTZ18Uplasmids. A 1kb DNA ladder was used.
The intensity of the components of the fluorescence that were parallel(IVV) and perpendicular(IVH) to the direction of thevertically polarized excitation light was determined by measuring the emittedlight through polarizers that were oriented vertically and horizontally. Thesteady-state anisotropy is given by:
(1)
where G is a grating correction factor for themonochromator’s transmission efficiency for vertically and horizontallypolarized light. This value is given by the ratio of the fluorescenceintensities of the vertical (IHV)to horizontal (IHH) components whenthe exciting light is polarized in the horizontal direction.
FD intensity and anisotropy decay measurements
Measurements were performed with the instruments that were describedpreviously (Lakowicz and Maliwal, 1985)and modified with a data acquisition card from ISS, Inc. (Urbana, USA) (Feddersen et al., 1989).The excitation source was a blue LED LNG992CFBW (Panasonic, Japan) with luminousintensity of 1,500 mcd. An LED driver LDX-3412 (ILX Lightwave, Boseman, USA)provided 30 mA of current at frequencies from 0.4 to 15 MHz. A 480 ± 20nm interference filter and a 630 nm cut-off filter were used for isolatingexcitation and emission, respectively. Rhodamine B in water (τ = 1.68 ns)was utilized as a lifetime standard. All of the measurements were performed at25°C.
The intensity decays were recovered from the FD data in terms of amultiexponential model:
(2)
where the preexponential factorai is the amplitude of eachcomponent, Σai = 1.0,τi is the decay time, andn is the number of exponential components. These valueswere determined by a nonlinear least squares analysis as described previously(Gratton et al.,1984; Lakowicz et al.,1984). Mean lifetimes were calculated by:
(3)
where fi is thefractional steady-state contribution of each component to the total emission,and Σfi is normalized tounity. fi is given by:
The best fits were obtained by minimizing values:
(5)
where ν is the number of degrees of freedom, andφω and mω arethe experimental phase and modulation, respectively. The subscriptc is used to indicate calculated values for assumed valuesof αi andτi, and δφ andδm are the experimental uncertainties.
The FD anisotropy decays were also analyzed in terms of themultiexponential model:
(6)
where gi is theamplitude of the anisotropy component with a rotational correlation timeθi,Σgi = 1.0, andr0 is the anisotropy in the absence ofrotational diffusion. The total anisotropy r0 was afitted parameter. The modulated anisotropy rωwas calculated by:
(7)
where Λω is the ratio of the amplitudesof the parallel and the perpendicular components of the modulated emission.
Results and Discussion
In this study, we demonstrated the usefulness of RuBD, a long-lifetime MLC,for probing the bending and torsional dynamics of supercoiled and relaxed pTZ18Uplasmids. Fig. 3 shows the emission spectra ofRuBD that is intercalated into the supercoiled and relaxed forms of the plasmids.For both forms, there was no difference in the emission spectra under ourexperimental conditions. RuBD showed an emission peak at about 620 nm. Steady-stateanisotropy measurements also showed no difference in the anisotropy values between 4and 45°C for both plasmid forms (Fig.4).
Fig. 3.
Emission spectra of RuBD intercalated into supercoiled and relaxedpTZ18U plasmids.
Fig. 4.
Temperature-dependent steady-state anisotropy of RuBD intercalated intosupercoiled and relaxed pTZ18U plasmids.
The FD intensity decays of RuBD intercalated into the supercoiled andrelaxed pTZ18U plasmids are shown in Fig. 5.The best fits of the intensity decay data were obtained using the triple exponentialmodel. The results are summarized in Table 1.The mean lifetime values for the supercoiled and relaxed plasmids were 148 and 160ns, respectively. The mean lifetime for the supercoiled plasmids was somewhatshorter than that for the relaxed plasmids. Because the interaction of water withthe nitrogen atoms on the dppz quenches the luminescence, this result suggests thatthe RuBD MLC was more efficiently shielded from water in the relaxed than in thesupercoiled plasmids. The observation of shorter lifetimes for the supercoiledplasmids agrees with our previous study that used supercoiled and linear pTZ18Uplasmids (Kang et al., 2002).The lifetime values of RuBD (Table 1) arelarger than reported previously (Lakowicz etal., 1995; Malak etal., 1997; Kang etal., 2002). The two earlier values (Lakowicz et al., 1995; Malak et al., 1997) were measured bytime-correlated single photon counting, which weights the shorter decay times moreheavily than does the FD measurements. In general, resolution of a tripleexponential decay is difficult (Lakowicz,1999), and the individual decay times are significantly uncertain.
Fig. 5.
Intensity decays of RuBD intercalated into supercoiled and relaxedpTZ18U plasmids. The symbols represent the measured phase and modulation values.The solid lines show the best multiexponential fits to the data. The middle andlower panels show plots of the residuals between the experimental data and thefitted curve.
Table 1.
Multiexponential intensity decay analyses of[Ru(bpy)2(dppz)]2+ (RuBD) intercalated intosupercoiled and relaxed pTZ18U plasmids.
| Plasmid Type | τi (ns) | αi | fia | <τ>a | b |
|---|---|---|---|---|---|
| Supercoiled | 272.6 | 0.09 | 0.40 | 148.0 | 2.2 |
| 70.9 | 0.49 | 0.55 | |||
| 7.7 | 0.42 | 0.05 | |||
| Relaxed | 291.2 | 0.10 | 0.41 | 159.5 | 1.5 |
| 74.3 | 0.51 | 0.54 | |||
| 9.7 | 0.39 | 0.05 |
Open in a separate window
aFractional intensities fi and meanlifetimes <τ> were calculated using Eqs. (4) and (3), respectively.
bThe values were calculated by Eq. (5), and the standard errors of phaseangle and modulation were set at 0.2° and 0.005, respectively.
In addition to the intensity decay measurements, the FD anisotropy decays ofRuBD intercalated into two forms of the pTZ18U plasmids were also measured (Fig. 6). The anisotropy decay data were best fitusing the two ≈ exponential model. The results are summarized in Table 2. The obtained rotational correlationtimes were 288 and 17 ns for the supercoiled, and 355 and 16 ns for the relaxedplasmids. The slow and fast rotational correlation times appear to be consistentwith the bending and torsional motions of the plasmids, respectively. Although thetorsional motions showed no significant difference between the two plasmid forms, weobserved somewhat shorter slow rotational correlation times for the supercoiledplasmids than the relaxed plasmids. This reflected a higher degree of bendingmotions in supercoiled than in relaxed plasmids. This result also agreed with ourobservations with the supercoiled and linear pTZ18U plasmids (Kang et al., 2002). The resolution ofthe differential phase values (Fig. 6, secondpanel) clearly showed that RuBD is a good probe for measuring the bending andtorsional dynamics of the supercoiled and relaxed pTZ18U plasmids. Fig. 6 also shows the modulated anisotropy values for RuBDthat was intercalated into both plasmid forms (Fig.6, first panel). We observed slightly lower modulated anisotropy valuesfor supercoiled plasmids. This is inconsistent with our steady-state anisotropydata, which showed no difference between the two forms (Fig. 4). Additionally, in our previous study (Kang et al., 2002), we observedslightly higher modulated anisotropy values for the supercoiled plasmids. Furtherexperimentation is required to understand this discrepancy.
Fig. 6.
Anisotropy decays of RuBD intercalated into supercoiled and relaxedpTZ18U plasmids. The symbols in the first and second panels represent thecalculated modulated anisotropy and the measured phase shift values,respectively. The solid lines show the best multiexponential fits to the data.The differential phase data in the second panel are based on the rotationalcorrelation times and amplitudes that are shown in Table 2. The superscripts S andR denote supercoiled and relaxed pTZ18U plasmids,respectively. The lower two panels show plots of the residuals between theexperimental data and fitted curve.
Table 2.
Multiexponential anisotropy decay analyses of[Ru(bpy)2(dppz)]2+ (RuBD) intercalated intosupercoiled and relaxed pTZ18U plasmids.
| Plasmid Type | θi | r0*g(i) | Σ(r0*g(i)) | a |
|---|---|---|---|---|
| Supercoiled | 288.1 | 0.057 | 0.122 | 3.2 |
| 16.5 | 0.065 | |||
| Relaxed | 354.6 | 0.062 | 0.135 | 4.5 |
| 15.9 | 0.073 |
Open in a separate window
aThe values were calculated by Eq. (5), and the standard errors of phaseangle and modulation were set at 0.2° and 0.005, respectively.
An important point of the present study is our use of a semiconductor lightsource for the FD intensity and anisotropy decay measurements. A variety of LEDs(including the high intensity UV, blue, and green LEDs) have been developed as aninexpensive and convenient light source. LEDs are easily modulated up to hundreds ofMHz without the need for a Pockels cell (Sipioret al., 1996). We believe that LEDs will become anideal light source for measuring the microsecond dynamics of biologicalmacromolecules.
The use of RuBD, a long-lifetime MLC, allowed us to determine the bendingmotions of the plasmids, which were not measurable using the fluorescence anisotropydecay of EB (< τ > ≈ 20 ns). However, as in the previousstudy (Kang et al., 2002), ithas to be pointed out that the lifetime of RuBD is far too short to measure theslower bending motions or end-over-end tumbling motions of the plasmids. The bendingmotions of DNA occur in about 100 ns to more than 100 μs (Schurr et al., 1992), which means thatthe time window of this report is near the lower end of bending dynamics of DNA. Thetime scale of end-over-end tumbling motions of plasmids spans from about 100μs to a few milliseconds, depending on the size of the plasmid (Langowski and Giesen, 1989; Langowski et al., 1992; Chirico and Baldini, 1996; Fishman and Patterson, 1996). Additionally, somewhat high probeconcentration was used because of the very low quantum yield (Q =0.008) of RuBD (Lakowicz et al.,2001). This may partially explain the small difference in the bending andtorsional dynamics between the supercoiled and relaxed plasmid forms, because thepresence of the intercalating agents unwinds the supercoils of closed circular DNA(Waring, 1970). The use of long-lifetimeMLCs to measure DNA dynamics is presently in its infancy. Additional MLCs withlonger lifetime, higher quantum yield, as well as long-wavelength absorption are yetto be developed.
Acknowledgments
This research was supported by the NIH, National Center for ResearchResources, RR 08119.
References
- Chirico G and Baldini G (1996) Rotational diffusion and internalmotions of circular DNA. II. Depolarized photon correlationspectroscopy. J. Chem. Phys104,6020–6026. [Google Scholar]
- DeGraff BA and Demas JN (1994) Direct measurement of rotationalcorrelation times of luminescent ruthenium (II) molecular probes bydifferential polarized phase fluorometry. J. Phys.Chem98,12478–12480. [Google Scholar]
- Feddersen BA, Piston DW and Gratton E (1989) Digital parallel acquisition infrequency domain fluorimetry. Rev. Sci.Instrum60,2929–2936. [Google Scholar]
- Fishman DM and Patterson GD (1996) Light scattering studies ofsupercoiled and nicked DNA. Biopolymers38,535–552. [PubMed] [Google Scholar]
- Friedman AE, Chambron J-C, Sauvage J-P, Turro NJ and Barton JK (1990) “Molecular lightswitch” for DNA: Ru(bpy)2(dppz)2+. J. Am.Chem. Soc112,4960–4962. [Google Scholar]
- Gratton E, Limkeman M, Lakowicz JR, Maliwal BP, Cherek H and Laczko G (1984) Resolution of mixtures offluorophores using variable-frequency phase and modulationdata. Biophys. J46,479–486. [PMC free article] [PubMed] [Google Scholar]
- Haugen GR and Lytle FE (1981) Quantitation of fluorophores insolution by pulsed laser excitation of time-filtereddetection. Anal. Chem53,1554–1559. [Google Scholar]
- Jenkin Y, Friedman AE, Turro NJ and Barton JK (1992) Characterization of dipyridophenazinecomplexes of ruthenium (II): The light switch effect as a function ofnucleic acid sequence and conformation.Biochemistry31,10809–10816. [PubMed] [Google Scholar]
- Kang JS, Abugo OO and Lakowicz JR (2002) Dynamics of supercoiled and linearpTZ18U plasmids observed with a long-lifetime metal-ligandcomplex. Biopolymers67, 121–128,2002. [PubMed] [Google Scholar]
- Kang JS and Lakowicz JR (2001) Fluorescence resonance energytransfer in calf thymus DNA from a long-lifetime metal-ligand complex tonile blue. J. Biochem. Mol. Biol34,551–558. [PMC free article] [PubMed] [Google Scholar]
- Kielkopf CL, Ding S, Kuhn P and Rees DC (2000) Conformational flexibility of B-DNAat 0.74 angstrom resolution: d(CCAGTACTGG)2. J. Mol.Biol296,787–801. [PubMed] [Google Scholar]
- Lakowicz JR (1999) Principles of FluorescenceSpectroscopy, 2nd ed., pp.98–101, KluwerAcademic/Plenum Publishers, New York, NewYork. [Google Scholar]
- Lakowicz JR, Gryczinski I, Piszczek G, Tolosa L, Nair R, Johnson ML and Nowaczyk K (2000) Microsecond dynamics of biologicalmolecules. Methods Enzymol.323,473–509. [PubMed] [Google Scholar]
- Lakowicz JR, Laczko G, Cherek H, Gratton E and Limkeman M (1984) Analysis of fluorescence decaykinetics from variable-frequency phase-shift and modulationdata. Biophys. J46,463–477. [PMC free article] [PubMed] [Google Scholar]
- Lakowicz JR, Malak H, Gryczynski I, Castellano FN and Meyer GJ (1995) DNA dynamics observed with longlifetime metal-ligand complexes.Biospectroscopy1,163–168. [Google Scholar]
- Lakowicz JR and Maliwal BP (1985) Construction and performance of avariable-frequency phase-modulation fluorometer.Biophys. Chem21,61–78. [PubMed] [Google Scholar]
- Lakowicz JR, Piszczek G and Kang JS (2001) On the possibility of long-wavelengthlong-lifetime high-quantum yield luminophores. Anal.Biochem288,62–75. [PMC free article] [PubMed] [Google Scholar]
- Langowski J and Giesen U (1989) Configurational and dynamicproperties of different length superhelical DNAs measured by dynamic lightscattering. Biophys. Chem34, 9–18. [PubMed] [Google Scholar]
- Langowski J, Kremer W and Kapp U (1992) Dynamic light scattering for study ofsolution conformation and dynamics of superhelical DNA.Methods Enzymol. 211,430–448. [PubMed] [Google Scholar]
- Lee JH and Choi BS (2000) Structure of a DNA duplex containinga site-specific Dewar isomer: structural influence of the 3’-T· G base pair of the Dewar product. J.Biochem. Mol. Biol33,268–275. [Google Scholar]
- Malak H, Gryczinski I, Lakowicz JR, Meyers GJ and Castellano FN (1997) Long-lifetime metal-ligand complexesas luminescent probes for DNA. J.Fluorescence7,107–112. [PMC free article] [PubMed] [Google Scholar]
- Murphy CJ and Barton JK (1993) Ruthenium complexes as luminescentreporters of DNA. Methods Enzymol.226,576–594. [PubMed] [Google Scholar]
- Okonogi TM, Alley SC, Reese AW, Hopkins PB and Robinson BH (2000) Sequence-dependent dynamics in duplexDNA. Biophys. J78,2560–2571. [PMC free article] [PubMed] [Google Scholar]
- Rajski SR, Jackson BA and Barton JK (2000) DNA repair: models for damage andmismatch recognition. Mutat. Res447,49–72. [PubMed] [Google Scholar]
- Schurr JM, Fujimoto BS, Wu P and Song L (1992) Fluorescence studies of nucleicacids: dynamics, rigidities, and structures; inTopics in Fluorescence Spectroscopy, Volume 3, BiochemicalApplications, Lakowicz JR (ed.), pp. 137–229,Plenum Press, New York, NewYork [Google Scholar]
- Sipior J, Carter JM, Lakowicz JR and Rao G (1996) Single quantum well light-emittingdiodes demonstrated as excitation sources for nanosecond phase modulationfluorescence lifetime measurements. Rev. Sci.Instr67,3795–3798. [Google Scholar]
- Terpetschnig E, Szmacinski H and Lakowicz JR (1997) Long-lifetime metal-ligand complexesas probes in biophysics and clinical chemistry.Methods Enzymol. 278,295–321. [PubMed] [Google Scholar]
- Waring M (1970) Variation of the supercoils in closedcircular DNA by binding of antibiotics and drugs: evidence for molecularmodels involving intercalation. J. Mol.Biol54,247–279. [PubMed] [Google Scholar]
