Transition on the entropic elasticity of DNA induced by intercalating molecules
aa r X i v : . [ phy s i c s . b i o - ph ] A p r Transition on the entropic elasticity of DNA induced byintercalating molecules
M. S. Rocha, M. C. Ferreira, and O. N. Mesquita
Departamento de F´ısica, ICEX, Universidade Federal de Minas Gerais,Caixa Postal 702, Belo Horizonte, CEP 31270-901, MG, Brazil (Dated: October 31, 2018)
Abstract
We use optical tweezers to perform stretching experiments on DNA molecules when interactingwith the drugs daunomycin and ethidium bromide, which intercalate the DNA molecule. Theseexperiments are performed in the low-force regime from zero up to 2 pN. Our results show thatthe persistence length of the DNA-drug complexes increases strongly as the drug concentrationincreases up to some critical value. Above this critical value, the persistence length decreasesabruptly and remains practically constant for larger drug concentrations. The contour length ofthe molecules increases monotonically and saturates as drugs concentration increases. Measured in-tercalants critical concentrations for the persistence length transition coincide with reported valuesfor the helix-coil transition of DNA-drug complexes, obtained from sedimentation experiments.
Key words:
DNA; daunomycin; ethidium bromide; persistence length; optical tweezers; singlemolecule
PACS numbers: 87.80.Cc, 87.14.Gg, 87.15.-v . INTRODUCTION DNA-drug interactions have been much studied along the past years. An importantmotivation for these studies is the fact that many of the studied drugs are used for treatmentof human diseases, particularly, in cancer chemotherapy.Single molecule stretching experiments using optical tweezers have yielded a great amountof information about DNA-protein and DNA-drug interactions [1, 2, 3, 4, 5, 6].Recently, we studied the interaction between psoralen and DNA when illuminated withultraviolet light A (UVA). Psoralen is a drug used to treat some skin diseases, like psoriasisand vitiligo. This drug intercalates the DNA molecule and can form covalent linkages withthe thymines if the complex is illuminated with ultraviolet light, modifying drastically itselasticity and impeding the DNA replication and transcription. The persistence length ofthe DNA-psoralen complexes formed after UVA illumination were measured in [7].Daunomycin and ethidium bromide (EtBr) are other examples of drugs which intercalatethe DNA molecule and can modify its elasticity, depending on the drug concentration. Bothdrugs unwind the DNA double helix when intercalating [8]. Daunomycin is an anthracy-cline antibiotic used in the treatment of various cancers. It inhibits DNA replication andtranscription when intercalating, impeding cell duplication [9]. Ethidium bromide (EtBr)is commonly used as a non-radioactive marker for identifying and visualizing nucleic acidbands in electrophoresis and in other methods of nucleic acid separation.Several works have reported different results for the effects of these drugs on the entropicelasticity of DNA molecules. In those works, the measured parameter used to study elasticitymodifications is the persistence length of the DNA-drug complex. Smith et al. [2] reportthat ethidium bromide does not modify the elasticity of the DNA molecule, only increasingits contour length by ∼ et al. [10] report that ethidium bromide causes a largeincrease in the contour length and a decrease in the persistence length of the complex for1 µ M of the drug, and at lower concentrations, an increase in both persistence and contourlengths. Recently, Sischka et al. [11] report the value of 28.1 nm for the persistence length ofDNA-daunomycin complexes and 20.7 nm for DNA-EtBr complexes, smaller than the bareDNA persistence length of about 50 nm. The authors have used in this work a concentrationof 1 µ M for both drugs, and a DNA concentration of 15 pM. In the present work, in order toclearly establish the effect of these intercalating drugs on the persistence length of the DNA2omplexes, we performed stretching experiments at various drugs concentrations, from zeroup to saturation of the complexes. We show that the values obtained for the persistencelength depend strongly on the concentration ratio between drug and DNA base pairs. Ourresults show that the persistence length of the complexes increases as we increase the drugconcentration until certain critical concentration is reached. Above this critical concentrationthe persistence length decreases abruptly and remains practically constant for larger drugconcentrations.
II. EXPERIMENTAL PROCEDURE
To measure the persistence and contour length of DNA molecules and DNA-drug com-plexes, we use optical tweezers [1, 2, 3, 4, 5, 6] and intensity autocorrelation spectroscopy[12].The samples consist of λ -DNA molecules in a PBS pH 7.4 with [NaCl] = 140 mM solution.We attach one end of the molecule to a microscope coverslip, and the other end is attachedto a polystyrene bead. To do this, we use the procedure described in [13]. We add the drugin the sample immediately before the measurements. The DNA concentration used in allexperiments was C DNA = 6.81 µ g/mL, which corresponds to a base pairs concentration of C bp = 11 µ M.Our optical tweezers is mounted in a Nikon TE300 microscope with an infinite correctedobjective (100X, N.A. = 1.4). The trapping laser is an infrared (IR) laser with λ = 832 nm(SDL, 5422-H1). The optical tweezers is used to trap the polystyrene bead attached to theend of the DNA molecule, so we can manipulate and stretch the DNA molecule.In addition, we use a He-Ne laser ( λ = 632.8 nm) as the scattering probe. The backscat-tered light by the polystyrene bead is collected by a photodetector, which delivers pulsesto a digital correlator. We, then, obtain the autocorrelation function of the backscatteredlight, which allows us to determine the stiffness of the optical trap, due to the Brownianmotion of the trapped bead.The next step in the experimental procedure is to obtain the force versus extension curvesfor the DNA molecules and DNA-drug complexes. To do this, we use the optical tweezersto trap the bead with the DNA while pulling the microscope slide, stretching the DNA.The backscattered light is collected while stretching the DNA. From the backscattered light3ntensity one obtains the displacement of the trapped bead in relation to its equilibriumposition, and by multiplying it times the tweezers’ stiffness, the force exerted by the DNAmolecule while it is stretched is obtained. The details about our experimental setup andexperimental procedure can be found in [7, 12].Finally, with the force versus extension curves, we use the approximate expression derivedby Marko and Siggia (Eq. 1) [14] to obtain the persistence and contour length of the DNAmolecules and DNA-drug complexes, F = k B TA zL + 14 (cid:16) − zL (cid:17) − , (1)where z is the DNA molecule end-to-end distance, k B is the Boltzmann constant, T is theabsolute temperature, A is the DNA persistence length and L is the DNA contour length.Figure 1 is a typical force versus extension curve obtained with this procedure, for adrug-free DNA molecule. Fitting this curve with Eq. 1, we extract the persistence length( A = 50 ± L = 16.5 ± µ m) for the λ -DNA. These valuescorrespond to well-known values reported in the literature [3, 15, 16]. III. RESULTS AND DISCUSSION
In this section we show the results obtained for the two drugs used: daunomycin andethidium bromide.
A. Daunomycin
We have performed experiments with DNA-daunomycin complexes for several drug con-centrations. In Fig. 2 we show the persistence length ( A ) of the complexes as a functionof total daunomycin concentration ( C D ) for fixed DNA base pairs concentration of C bp =11 µ M. We denote by C D the total daunomycin concentration used to prepare the sample,which is the sum of both the bounded to DNA and the free drug concentration in solution.The point which the drug concentration is zero corresponds to the drug-free DNA situa-tion with A = 50 nm.The behavior of the persistence length A as a function of daunomycin concentration C D can be described as follows: it initially increases with C D until it reaches a maximum value4 ∼
280 nm) at the critical concentration C criticalD = 18.3 µ M. Then, it decays abruptly toaround 75 nm and remains practically constant at this value even if we continue to increase C D . The contour length increases monotonically from 16.5 ± µ m up to the saturationvalue of 21 ± µ m. These mean values are obtained performing an average over manydifferent DNA molecules and DNA-drug complexes. Such distribution of contour lengthvalues was also observed by Mihailovic et al. [17].Also, we can estimate the exclusion parameter n (number of total base pairs divided bythe number of total intercalated drug molecules) from our experimental data. The averagevalue of the contour length for DNA-daunomycin complexes obtained when using a saturatedconcentration of the drug increases about 27% relative to drug-free DNA contour length(16.5 µ m). This means that when all possible drug molecules are intercalated, the DNAincreases its contour length by 4.5 µ m. Knowing that each intercalated daunomycin moleculeincreases the contour length of the complex by 0.31 nm [8], we determine the total number ofintercalated daunomycin molecules, which is around 14500. Finally, the exclusion parametercan be obtained by dividing the number of base pairs of the λ -DNA (48500) by the numberof total intercalated drug molecules (14500). We obtain n = 3.3, in good agreement withthe value 3.5 reported in [9].For comparison purposes, Fig. 3 shows two force versus extension curves (normalized bythe contour length) for two daunomycin concentrations, before and after the transition. Thedata points in this figure are smoothed for better visualization, i. e. , Brownian fluctuationsare averaged out. B. Ethidium Bromide (EtBr)
The behavior of the persistence length as a function of the drug concentration for DNA-EtBr complexes is very similar to the DNA-daunomycin complexes. The difference is that inthis case the transition occurs at a lower drug concentration (see Fig. 4) for the same DNAbase pairs concentration C bp = 11 µ M. The maximum value measured for the persistencelength of DNA-EtBr complexes is ∼
150 nm, at the critical concentration C criticalE = 3.1 µ M.The contour length increases monotonically from 16.5 ± µ m up to the saturation valueof 23 ± µ m. Again, these values for the contour lengths are averages over many differentmolecules. 5epeating the same calculation for the exclusion parameter of EtBr, which increasesthe DNA contour length by 0.34 nm per intercalated molecule [11], we obtain n = 2.5, inreasonable agreement with the value 2.01 reported in [18]. C. Equilibrium binding constants
In our experiments we control the total drug concentration C T and the total concentrationof DNA base pairs C bp . To discuss the elastic properties of the DNA complex formed theimportant parameter to consider is the ratio r between the concentration of bounded drug( C b ) per concentration of DNA base pairs ( C bp ). In order to obtain r , the binding of moleculesto DNA is analyzed using the neighbor exclusion model [9]. A closed form for this modelwas obtained by McGhee and von Hippel [19] and can be expressed by the equation rC f = K i (1 − nr ) " − nr − ( n − r n − , (2)where r is ratio between the concentration of bounded drug ( C b ) per concentration of DNAbase pairs ( C bp ), C f is the concentration of free drug (not bounded), K i is the intrinsic bind-ing constant and n is the exclusion parameter in base pairs. For a more detailed discussionabout the neighbor exclusion model, see [19].The concentration of free drug ( C f ) can be simply related with the concentration ofbounded drug ( C b ) and the total drug concentration ( C T ) through the equation C T = C f + C b . (3)Using Eq. 2 and 3 with the determined exclusion parameter ( n = 3.3), the intrinsicbinding constant reported in [9] for daunomycin, K i = 7 × M − , the critical daunomycinconcentration measured in this work ( C criticalD = 18.3 µ M), and the concentration of DNAbase pair used in our experiments ( C bp = 11 µ M), we can determine the critical ratio r c ,which we define as the value of r at the abrupt transition for the value of the persistencelength. We then obtain the value r c = 0.248.Similarly, for EtBr, we use the parameters n = 2.5, K i = 1.5 × M − [20], and C criticalE = 3.1 µ M determined again from the abrupt change in persistence length. We obtain r c =0.131. In Section III D we compare the values obtained for r c with those reported in theliterature for a sedimentation experiment. 6t is important to mention that K i varies with the ionic strength of the solution. Thevalues used here are the values for the ionic concentrations used in our experiments. D. Interpretation of the DNA-drug complexes elasticity results
For low drug concentrations, drug intercalation in the DNA molecule increases the rigidityof the complex (see Figs. 2 and 4). This is consistent with the results of Vladescu et al. [21],which shows that EtBr stabilizes the DNA double-helix for low drug concentrations. Theyhave performed melting experiments with various EtBr concentrations, from zero to 2.5 µ M,showing that EtBr intercalation stabilizes the DNA double-helix in this concentration range.Therefore, we expect an increase of the persistence length of DNA-drug complexes in thislow concentration range. Figure 4 shows this increase for EtBr, and Fig. 2 shows a similarresult for daunomycin.For high drug concentrations, i. e. , above the critical concentration (peak of Figs. 2 and4), the persistence length of the complexes decays abruptly and remains practically constant.It is well-known that intercalation unwinds the DNA double-helix [8]. Due to unwindingand above some drug critical concentration, the complexes can have a helix-coil transition,which can cause DNA denaturing as the DNA is stretched, decreasing the persistence lengthof DNA-drug complexes as seen in Figs. 2 and 4. The unwinding angle per intercalated EtBrdrug molecule is approximately 1.7 times greater than that for daunomycin intercalation [8].Therefore we expect that the transition occurs for EtBr at a lower drug concentration ascompared with daunomycin, if the same DNA concentration is used. This is confirmedexperimentally in our data of Figs. 2 and 4.Sedimentation experiments performed with circular DNA as a function of daunomycinand ethidium bromide concentrations display a minimum in the sedimentation coefficientS at r c = 0.192, for daunomycin and r c = 0.114 for ethidium bromide [8]. The minimum inthe sedimentation coefficient S is associated with a helix-coil transition, due to unwindingof the DNA double-helix by the intercalating drugs [8]. These numbers agree within 15 to30% with the values of r c determined from our DNA persistence length measurements. Thisindicates that the abrupt change of the DNA persistence length for both drugs might bealso caused by a helix-coil transition due to the unwinding of the DNA double-helix as thedrugs intercalate into it. The reasonable agreement between the critical ratios r c obtained7rom sedimentation experiments [8] and from our measurements of the persistence lengthtransition provides an evidence that a helix-coil transition is probably what we are observingin our experiments.In addition, it is known that EtBr (and also most intercalating drugs) exhibits multi-modality at their interaction with DNA [22, 23]. The type of interaction varies with thedrug concentration. The abrupt transition shown in Figs. 2 and 4 might as well be causedby different ways of drug binding to DNA. IV. CONCLUSION
We have made systematic measurements of the entropic elasticity variation of a λ -DNAmolecule when interacting with two drugs, daunomycin and ethidium bromide, as a functionof their concentrations. Our results show that the persistence length of the DNA-drugcomplexes increases strongly as the drug concentration increases, for low concentrations.Above certain critical drug concentration the persistence length decreases abruptly andremains practically constant for high drug concentrations. This behavior is quite similarfor both daunomycin and EtBr, as shown in Figs. 2 and 4. Our results suggests that theabrupt transition observed in the persistence length might be due to a helix-coil transitionand denaturing of DNA-drug complexes above the critical concentration, resulting in adecrease of the persistence length. V. ACKNOWLEDGEMENTS
This work was supported by the Brazilian agencies: Conselho Nacional de Desenvolvi-mento Cient´ıfico e Tecnol´ogico (CNPq), FAPEMIG, FINEP-PRONEX, Instituto do Milˆeniode Nanotecnologia e Instituto do Milˆenio de ´Optica N˜ao-linear e Biofotˆonica - MCT. M. S.R. acknowledges support by LNLS. [1] A. Ashkin, Phys. Rev. Lett. , 156& (1970).[2] S. B. Smith, L. Finzi, and C. Bustamante, Science , 1122 (1992).[3] M. D. Wang, H. Yin, R. Landick, J. Gelles, and S. M. Block, Biophys. J. , 1335 (1997).
4] G. V. Shivashankar, M. Feingold, O. Krichevsky, and A. Libchaber, Proc. Natl. Acad. Sci.USA , 7916 (1999).[5] C. G. Baumann, V. A. Bloomfield, S. B. Smith, C. Bustamante, M. D. Wang, and S. M.Block, Biophys. J. , 1965 (2000).[6] D. G. Grier, Nature , 810 (2003).[7] M. S. Rocha, N. B. Viana, and O. N. Mesquita, J. Chem. Phys. , 9679 (2004).[8] H. Fritzsche, H. Triebel, J. B. Chaires, N. Dattagupta, and D. M. Crothers, Biochemistry ,3940 (1982).[9] J. B. Chaires, N. Dattagupta, and D. M. Crothers, Biochemistry , 3933 (1982).[10] I. Tessmer, C. G. Baumann, G. M. Skinner, J. E. Molloy, J. G. Hoggett, S. J. B. Tendler, andS. Allen, Journal of Modern Optics , 1627 (2003).[11] A. Sischka, K. Toensing, R. Eckel, S. D. Wilking, N. Sewald, R. Rios, and D. Anselmetti,Biophys. J. , 404 (2005).[12] N. B. Viana, R. T. S. Freire, and O. N. Mesquita, Phys. Rev. E , Art. No. 041921 (2002).[13] G. V. Shivashankar, G. Stolovitzky, and A. J. Libchaber, Appl. Phys. Lett. , 291 (1998).[14] J. F. Marko and E. D. Siggia, Macromolecules , 8759 (1995).[15] T. Strick, J. F. Allemand, V. Croquette, and D. Bensimon, Prog. Biophys. Mol. Biol. , 115(2000).[16] J. C. Neto, R. Dickman, and O. N. Mesquita, Physica A , 173 (2005).[17] A. Mihailovic, I. Vladescu, M. McCauley, E. Ly, M. C. Willians, E. M. Spain, and M. E.Nunez, Langmuir , 4699 (2006).[18] J. E. Coury, L. McFail-Isom, L. D. Williams, and L. A. Bottomley, Proc. Natl. Acad. Sci.USA , 12283 (1996).[19] J. D. McGhee and P. H. von Hippel, J. Mol. Biol. , 469 (1974).[20] B. Gaugain, J. Barbet, N. Capelle, B. P. Roques, and J. L. Pecq, Biochemistry , 5078(1978).[21] I. D. Vladescu, M. J. McCauley, I. Rouzina, and M. C. Willians, Phys. Rev. Lett. , 158102(2005).[22] P. O. Vardevanyan, A. P. Antonyan, G. A. Manukyan, and A. T. Karapetyan, Exp. Mol. Med. , 205 (2001).[23] A. Arabzadeh, S. Z. Bathaie, H. Farsam, M. Amanlou, A. A. Saboury, A. Shockravi, and . A. Moosavi-Movahedi, Int. J. Pharmac. , 47 (2002). igure Legends Figure 1.
Force as a function of extension for a drug-free DNA molecule. By fitting this curve withEq. 1, we determine the persistence length A = 50 ± L = 16.5 ± µ m. Figure 2.
Persistence length A of DNA-daunomycin complexes as a function of drug concentrationfor fixed DNA concentration ( C bp = 11 µ M). A initially increases with C D until it reachesa maximum value ( ∼
280 nm) at the critical concentration C criticalD = 18.3 µ M. Then, thepersistence length decays abruptly to around 75 nm and remains practically constant at thisvalue even if we continue to increase the drug concentration.
Figure 3.
Force versus extension curves (normalized by the contour length) for two daunomycinconcentrations. The data Brownian fluctuations are averaged out for better visualization.
Circles : C D = 20.1 µ M (above the critical concentration) and A ∼
61 nm; triangles : C D =18.3 µ M and A ∼
263 nm. Dashed lines are fittings using Eq. 1.
Figure 4.
Persistence length of DNA-EtBr complexes as a function of drug concentration for fixedDNA concentration ( C bp = 11 µ M). Here, the transition occurs at a lower drug concentration.The maximum value measured for the persistence length of DNA-EtBr complexes is ∼ C criticalE = 3.1 µ M.11