Salt Has a Biphasic Effect on the Higher-Order Structure of a DNA

ARTICLE pubs.acs.org/JPCB

Salt Has a Biphasic Effect on the Higher-Order Structure of a DNAProtamine Complex Naoko Makita,† Yuko Yoshikawa,*,‡ Yoshiko Takenaka,§ Takahiro Sakaue,||,^ Mari Suzuki,#,r Chika Watanabe,# Tamotsu Kanai,O Toshio Kanbe,[ Tadayuki Imanaka,‡ and Kenichi Yoshikawa#,r †

Faculty of Environmental and Information Sciences, Yokkaichi University, Yokkaichi 512-8512, Japan Laboratory of Environmental Biotechnology, Research Organization of Science and Engineering, Ritsumeikan University, Kusatsu 525-8577, Japan § Flucto-Order Functions Team, RIKEN Advanced Science Institute, Wako, 351-0198, Japan Department of Physics, Kyushu University 33, Fukuoka 812-8581, Japan ^ PRESTO, JST, Kawaguchi, 332-0012, Japan # Department of Physics, Graduate School of Science, Kyoto University, Kyoto 606-8502, Japan r Spatiotemporal Order Project, ICORP, JST, Kyoto University, Kyoto 606-8502, Japan O Department of Synthetic Chemistry and Biological Chemistry, Graduate School of Engineering, Kyoto University, Kyoto 606-8510, Japan [ Laboratory of Medical Mycology, Research Institute for Disease Mechanism and Control, School of Medicine, Nagoya University, Nagoya 464-0064, Japan

)

‡

ABSTRACT: We observed single DNA molecules by fluorescence microscopy to clarify the effect of protamine on their higher-order structure. With an increase in the protamine concentration, the conformation of DNA molecules changes from an elongated coil state to a compact state through an intermediate state. Furthermore, the long-axis length of DNA gradually decreases while maintaining a distribution profile with a single peak. Such behavior is markedly different from the conformational transition of DNA induced by small polyamines such as spermidine and spermine, where individual DNA molecules exhibit an all-or-none transition from a coil to a globule state and the size distribution is characterized by twin peaks around the transition region. Next, we examined the effect of salt on the conformation of the DNAprotamine complex. Interestingly, at a fixed concentration of protamine, DNA tends to shrink with an increase in the NaCl concentration up to 300 mM, and then swells with a further increase in the NaCl concentration, that is, biphasic behavior is generated depending on the salt concentration. For comparison, we examined the effect of salt on DNA compaction induced by the trivalent polyamine spermidine. We confirmed that salt always has an inhibitory effect on spermine-induced compaction. To clarify this biphasic effect of salt on protamine-induced DNA compaction, we performed a numerical simulation on a negatively charged semiflexible polyelectrolyte in the presence of polycations with relatively large numbers of positive charges by taking into account the effect of salt at different concentrations. The results showed that salt promotes compaction up to a certain concentration and then tends to unfold the polyelectrolyte chain, which reproduced the experimental observation in a semiquantitative manner. This biphasic effect is discussed in relation to the specific shielding effect that depends on the salt concentration.

’ INTRODUCTION In vitro studies on the dynamics of DNA compaction and condensation can provide useful insights for understanding DNA packaging and its mechanistic characteristics in living cells.119 Numerous in vitro studies have reported that DNA compaction is induced by various condensing agents, including polyamines,49 metal compounds,10,11 neutral polymers,12,13 cationic surfactants,14 polyampholyte,15 and basic proteins.1618 Furthermore, compact DNAs show various morphologies, such as toroid, rod, racket, and spool-like structures.10,19 r 2011 American Chemical Society

We have performed direct observations of the higher-order structure of single DNA molecules in solution using fluorescence microscopy and found that long duplex DNA molecules with a size of larger than several tens of kilo base-pairs exhibit a discrete conformational transition from an elongated coil state to a folded compact state upon the addition of various condensing agents, Received: November 29, 2010 Revised: February 3, 2011 Published: March 29, 2011 4453

dx.doi.org/10.1021/jp111331q | J. Phys. Chem. B 2011, 115, 4453–4459

The Journal of Physical Chemistry B whereas short DNA fragments behave like rigid rods and cannot undergo such a discrete folding transition.20 Since a genomic DNA molecule is a very long anionic polymer, studies on the nature of the conformational change of long DNA induced by condensing agents should serve as a simple model for understanding the change in the higher-order structure of genomic DNA in relation to its biological function. In the present article, we focus on the interaction of large DNA with salmon protamine, a small basic protein containing 21 positively charged arginines throughout 32 amino acids.21 In sperm nuclei of many vertebrates, a long DNA is packaged by protamine into a highly compact, biologically inactive form of chromatin. The manner of packaging plays a key role in spermatogenesis. In vitro studies using electron and atomic force microscopy,16,2225 and light scattering,26,27 have investigated how protamine compacts DNA. It has been shown that DNA molecules are comprised of multiple doughnut-shaped toroidal structures with a diameter of around 60 nm, where an individual toroid is made from 50 to 60 kb segments of DNA. Brewer et al. observed the folding/unfolding process of single DNA molecules interacting with protamine by applying an optical laser manipulation, where one end of DNA labeled with biotin is attached to a streptavidine-coated bead and the bead is optically trapped.21 They found that compaction starts from the end of the DNA molecule and the compact part grows at an almost constant speed. In this study, we used a fluorescence microscopic technique to investigate protamine-induced conformational changes in DNA in solution at the single-molecule level without any chemical modification of the structure itself. Because the interaction of DNA with protamine is due to electrostatic attraction between the negatively charged DNA strand and positively charged amino acids on the protein, the conformation of the resulting complex is expected to be sensitive to the concentration of coexisting salt. In our observations over a wide range of salt concentrations (1 mM to 1 M), we found that salt has a unique biphasic effect on protamine-induced DNA compaction.

’ EXPERIMENTAL METHODS Materials. T4 phage DNA (166 kbp, 57 μm) was purchased from Nippon Gene Co., LTD (Toyama, Japan). Salmon protamine-HCl (21 positively charged arginine residues) was obtained from Sigma (St. Louis, MO, USA). YOYO-1 (quinolinium, 1,10 -[1,3-propanediyl-bis[(dimethyliminio)-3,1-propanediyl]]bis[4-[(3-methyl-2(3H)-benzoxazolylidene)methyl]]-tetraiodide) was obtained from Molecular Probes Inc. (Eugene, OR, USA) and DAPI (4,6-diamidino-2-phenylin-dole) was obtained from Dojindo Molecular Technologies, Inc. (Kumamoto, Japan). The antioxidant 2-mercaptoethanol (2-ME) and other chemicals were purchased from Wako Pure Chemical Industries (Osaka, Japan). Direct Observation of Protamine-Induced DNA Conformation in Solution by Fluorescence Microscopy. To visualize DNA molecules, two different fluorescent dyes, DAPI and YOYO-1, were used to stain DNA. For DAPI staining, T4 phage DNA at a final concentration of 0.3 μM was dissolved in 1 mM NaCl containing different concentrations of protamine and 0.3 μM of DAPI. For the observation with YOYO-1 staining, T4 phage DNA was dissolved in a mixture of 10 mM Tris-HCl buffer, pH 7.5, with different concentrations of protamine. The fluorescent dye YOYO-1 was added to the sample solution at

ARTICLE

0.1 μM with 4% (v/v) 2-mercaptoethanol. Fluorescence images of DNA molecules were obtained using an Axiovert 135 TV (Carl Zeiss, Jena, Germany) microscope equipped with an oilimmersed 100 objective lens and recorded on videotape through an EBCCD camera (Hamamatsu Photonics, Hamamatsu, Japan). All observations were carried out at room temperature (∼20 C). Measurement of the Effect of Salt on a DNAProtamine Complex. To measure the conformation of DNA under a NaCl concentration of up to 1 M, we used DAPI as the fluorescence dye instead of YOYO-1. It was reported that high salt concentrations inhibit the binding of YOYO-1 to DNA.28

’ RESULTS AND DISCUSSION Direct Observation of the Conformational Change in DNA Induced by Protamine. Figure 1 exemplifies fluorescence

microscopic images of DNA molecules moving freely in aqueous solution, where DNAs were observed with (A) DAPI in 1 mM NaCl solution and with (B) YOYO-1 in 10 mM Tris-HCl buffer solution. To obtain fluorescence microscopic images with higher resolution, we also observed DNA molecules fixed on a glass surface (right of part B of Figure 1). On the right of each fluorescence image, a quasi-3D picture is shown to provide the spatial fluorescence intensity distribution. Without protamine, individual DNA molecules exist in an elongated coil state (top pictures in Figure 1). With the addition of protamine (0.02 μg/mL in (A) and 0.05 μg/mL in (B)), a partially folded conformation appears, where the compact and elongated parts coexist in a single DNA molecule, as shown in the middle pictures in Figure 1. The pictures at the bottom of Figure 1 show fluorescence images of fully folded compact DNA in the presence of increased amounts of protamine: 0.05 μg/mL in (A) and 0.3 μg/mL in (B). Thus, it is clear that, for the conditions in both (A) and (B) in Figure 1, DNA molecules exhibit a conformational change from an elongated coil state to a compact state through a partial globule as an intermediate state. A difference between these two experiments is that more protamine is necessary for (B) because of the inhibitory effect of the intercalating agent on compaction. In the present study, we examined the effect of salt mainly by using the experimental condition as in (A), that is, DNA was stained by DAPI. This is because it was difficult to observe single DNA molecules at high salt concentrations with YOYO-1, that is, DNA molecules become invisible by fluorescence microscopy. Next, we compared the effect of protamine on the higherorder structure of DNA with that of spermidine(3þ) at a fixed low-salt condition. Figure 2 shows the distribution of the longaxis length of DNA, where the conformation is classified as elongated coil, partial globule, or globule. From the timedependent fluctuation of the fluorescence image, it was possible to distinguish the chain conformation without difficulty. It is clear that both protamine and spermidine cause the folding transition from a coil to a compact state with an increase in their concentrations. However, there is a marked difference on the manner of the folding transition. With protamine, partially globule DNA appears as indicated by the gray bar in part A of Figure 2. In contrast, with spermine, individual DNA molecules undergo an on/off-type discrete transition between the coil and compact states. An all-or-none-type transition of DNA has frequently been reported for compaction induced by polycations with a charge number of around 3 to 5. With an increase in the charge number 4454

dx.doi.org/10.1021/jp111331q |J. Phys. Chem. B 2011, 115, 4453–4459

The Journal of Physical Chemistry B

ARTICLE

Figure 1. Fluorescence microscopic observation of individual T4 DNA molecules stained with DAPI (A) and YOYO-1 (B). (A) Left: Fluorescent images of DNA molecules stained with DAPI in 1 mM NaCl solution: (a) an elongated coil state at 0 μg/mL protamine, (b) a partially folded state at 0.02 μg/mL protamine, and (c) a compact folded state at 0.05 μg/mL protamine. Middle: Quasi-3D pictures of the fluorescence intensity. Right: Schematic representation of fluorescent images of DNA . (B) Left: Fluorescent images of DNA molecules stained with YOYO-1 in 10 mM Tris-HCl buffer solution: (d) an elongated coil state at 0 μg/mL, (e) a partially folded state at 0.05 μg/mL protamine, and (f) a compact folded state at 0.3 μg/mL protamine. Middle: Quasi-3D pictures of the fluorescence intensity. Right: Fluorescent images of DNA molecules on a glass surface.

to on the order of several tens, the profile of the folding transition becomes continuous. The total positive charge of polycation required for DNA compaction decreases with an increase in the charge number of the polycation. The observed experimental trend in Figure 2 corresponds to the above-mentioned general trend that has been reported so far. In relation to this, Haaf and Ward observed a beaded structure on highly extended chromatin fibers released from mature human sperm by fluorescence microscopy.29 The images in part e of Figure 1 correspond well to those seen in native sperm chromatin. Effect of the NaCl Concentration on Protamine-Induced DNA Compaction. To understand the general trend in the nature of the conformational change of DNA in the presence of both protamine and coexisting salt, we performed fluorescence microscopic observations of DNA molecules while varying the concentrations of protamine and NaCl. Part A of Figure 3 shows the change in the distribution of the long-axis length at different concentrations of protamine and NaCl. It is clear that protamine-induced DNA compaction is promoted by NaCl up to a concentration of 300 mM. In contrast, DNA molecules assumed an elongated coil conformation at a NaCl concentration above 500 mM. Thus, DNAprotamine interaction exhibits biphasic behavior that depends on the NaCl concentration. Part B of Figure 3 shows the changes in the long-axis length depending on the NaCl concentration fixed protamine concentrations. The biphasic effect of salt is demonstrated. Note that the NaCl concentration at the minimum long-axis length corresponds to a physiological salt concentration. On the basis of these measurements, we created a phase diagram of the conformational change of DNA as a function of the NaCl and protamine concentrations (part A of Figure 4). For

comparison, we also show the phase diagram for the spermidineinduced folding transition of DNA. With spermidine, salt always tends to unfold DNA into a coil state, and thus the DNA conformation shows a monotonic change depending on the salt concentration. In contrast, protamine causes a tightly compact state at salt concentrations of around 100300 mM (part A of Figure 4). With an increase in the NaCl concentration above 500 mM, and with a decrease below this NaCl concentration, DNA molecules tend to exhibit an elongated coil conformation. Vilfran et al. reported that no salmon protamine-DNA condensates were observed at 1 M NaCl using transmission electron microscopy and light scattering experiments.24 Zlatanova et al. also observed a similar effect of salt with somatic chromatin: the chromatin fiber is extended at a low ionic strength and becomes more condensed at physiological salt concentrations.30 Such a biphasic effect of salt was also observed in histone proteinDNA interactions.31 Theoretical Consideration of the Effect of Salt on DNA Protamine Interaction. In our experiment, we found that salt had a biphasic effect on the protamine-induced compaction of DNA. With regard to unfolding at high salt concentrations, we can consider the change in protamine binding equilibrium toward the release from DNA. It has been well established that the interaction between oppositely charged polyelectrolytes becomes less favorable with an increase in salt in the bulk solution. This can be interpreted by taking into account the effect of the release of counterions into the bulk solution, accompanied by the association between the oppositely charged polyelectrolytes. The increase in translational entropy upon the release of the counterions decreases with an increase in the salt concentration in the solution. Another explanation for this salt 4455



ARTICLE

Figure 2. Distributions of long-axis length of DNA together with an assignment of conformational characteristics in DNA images: coil (white bar), partial globule (gray bar) and compact globule (black bar) states of DNA molecules in 1 mM NaCl solution at various concentrations of protamine (A) and spermidine (B).

effect is that it is due to electric shielding by small counterions. Thus, this enhanced shielding at high salt concentrations shifts the binding equilibrium between the oppositely charged polyelectrolytes toward dissociation. Note that both explanations (change in the translational entropy of counterions and the shielding effect) predict the same trend of weakening of the association with an increase in the salt concentration. However, the salt effect at low concentrations is attributable to the electrostatic instability of DNA complexed with protamine. When the charges of the polyanion and polycation do not match each other exactly as in the experimental condition in this study, we have to consider the effect of the surviving electric charge along the DNA chain. It is natural to expect that the complex will exhibit large electric instability at lower salt concentrations. Thus, we can expect a swollen conformation at a low salt concentration and a compact conformation at a higher salt concentration. Such an electrostatic instability becomes less important when the number of positive charges per single polycation molecule becomes smaller by decreasing the electrostatic attraction between the polycation and DNA. The large difference between the action of protamine and spermidine on DNA at low salt concentrations is, thus, attributable to the difference on the effect of electrostatic interaction. To examine this expectation regarding the biphasic effect salt, we performed numerical simulations on a simple model system by mimicking the essence of the interaction between DNA and protamine. In a recent article, we showed that the complexion behaviors of a long DNA chain with oppositely charged nanoparticles could be well-described by a coarse-grained model composed of a semiflexible polyelectrolyte and oppositely

Figure 3. (A) Distributions of the long-axis length of T4 DNA molecules at various concentrations of protamine and NaCl. Black bar: a compact globule state, White bar: coil and partial globule states. (B) Long-axis length as a function of the NaCl concentration. The lines in the figure are provided simply as a guide.

charged nanospheres.32 We adopted essentially the same model here. We prepared a DNA molecule modeled by Nm spherical monomers of diameter σm and charge zm (in units of the elementary charge) and protamine molecules modeled by Np spherical molecules of diameter σp and charge zp in a periodic cubic box of size L = 100σm. For a DNA molecule, adjoining monomers along the chain are connected by the harmonic bonding potential Ubond ¼

∑

kbond ðjri ri þ 1 j σm Þ2 2σ 2m chain

where ri is the coordinate of the ith monomer of a DNA molecule. We chose a relatively large spring constant kbond = 400 to keep 4456



ARTICLE

salt Csalt as κ = (8πlBCsalt)1/2 and the exponent of the Morse potential R = 24. Note that, even though the DebyeH€uckel approximation shows a serious limitation for such a complex with a highly charged polyelectrolyte, this type of interaction is known to reproduce the complexion behaviors of oppositely charged objects even under conditions of counterion condensation with strong electrostatic coupling.32 The underdamped Langevin equation was used as the equation of motion for the ith unit of a DNA monomer or a protamine molecule: mI

Figure 4. Phase diagrams of the DNA conformation as a function of the (A) NaCl and protamine concentrations and (B) NaCl and spermidine concentrations. The lines represent the region for the appearance of globule DNA.

the bond length at a nearly constant value σm, by considering the mechanical stability of double-stranded DNA with respect to stretching deformation. The mechanical stiffness of a DNA chain as a semiflexible polymer can be interpreted by the following bending potential Ubend: " #2 ðB ri B r i 1Þ 3 ð B r iþ1 B r iÞ Ubend ¼ kbend 1 σ2m chain

∑

where we set the bending parameter kbond = 10 so that the mechanical persistence length is lp = 12σm. All of the particles and monomeric units interact through electrostatic forces as well as steric forces because of their excluded volumes (here modeled by the repulsive part of the Morse potential). Electrostatic parts are incorporated through the linearized DebyeH€uckel potential with the inclusion of the finite size effect of particles. Therefore, the interaction energy UIJ between units I and J (where the label I or J refers to either a DNA monomer or a protamine molecule) separated by distance r is UIJ ¼

lB zI zJ exp½kðr ðσI þ σ J Þ=2Þ ð1 þ kσ I =2Þð1 þ kσJ =2Þ r þ exp½Rðr ðσI þ σJ Þ=2Þ

where the Bjerrum length lB = 0.5σm and the inverse Debye screening length κ are tuned by the concentration of monovalent

d2 rI, i drI, i DU þ RI, i ðtÞ ¼ γI DrI, i dt 2 dt

where mI, rI,i and γI are the mass, the coordinate, and the friction constant of a monomer of DNA (I = m) or a protamine molecule (I = p), respectively. The internal energy is expressed as U = Ubond þ Ubend þ (1/2)ΣI6¼JUIJ with units of thermal energy kBT. Random force RI,i(t) is Gaussian white noise that obeys the fluctuationdissipation theorem. The ratio of the friction constant between the monomer γm and the core particle γp was evaluated according to Stokes law. We chose a relatively large mass mm = mp(σm/σp)3 = 1 to save calculation time, but all of this setup associated with the inertial part should not affect the motion within the time scale of interest, which is much longer than the relaxation time of velocity. The dynamics of the system were evaluated using a leapfrog algorithm with a time-step of Δt = 0.005τ, where τ = γmσ2m/kBT is a unit time-step. We started from random initial configurations, and performed all simulations for typically more than 105 time-steps, which allowed us to obtain good statistics. We fixed the parameters Nm = 200, σm = 1, zm = 5, σp and zp = 40 and performed the simulation by changing the number of protamine molecules Np and the salt concentration κ. The total charge of a DNA and protamine molecules in the system is denoted as Q(sys) = zmNm þ zpNp. The electrostatic interaction between a DNA and protamine molecules is strong, and thus, once a protamine molecule diffuses near a DNA and contacts it, complexion, that is wrapping of a DNA chain around the protamine molecule, occurs rapidly. This wrapping is irreversible in the sense that the wrapped protamine molecules never dissociate. Such a wrapping takes place one by one when a free protamine molecule diffuses near a DNA chain and eventually all of the protamine molecules are involved in the complex with a single DNA. Optimum complex structures would then be searched in the course of thermal fluctuations. The net charge Q* of this complex is an important quantity that is equal to the total charge Q* = Q(sys), that is there are no free protamine molecules in the bulk solution under our parameter conditions. Figure 5 shows the radius of gyration Rg of a DNA chain as a function of the monovalent-salt concentration with different numbers of protamine molecules (Np = 8,14,20). We can identify the following characteristics: (i) DNA assumes compact conformations around the optimum salt concentration Csalt = C(opt) salt ≈ 0.88/σ3m. (ii) The degree of compactness depends on Np, and thus, Q*. (iii) DNA expands at a higher salt concentration Csalt > C(opt) salt , and is comparable in size to that without protamine molecules (Np = 0). (iv) At a lower salt concentration Csalt < C(opt) salt , the size of a DNA chain may also increase depending on the value of Q*. Complexes ad of part B of Figure 5 show typical snapshots of the complex for various salt concentrations in the case of Np = 14. 4457



Figure 5. (A) Radius of gyration of a DNA chain as a function of the monovalent-salt concentration. The number of protamine molecules Np is 8 (plus, þ), 14 (open diamond, )), and 20 (open square, 0). The lines in the figure are provided simply as a guide. (B) Snapshots of the complex for various salt concentrations in the case of Np = 14 are shown in (a)(d). Salt concentrations (Debye lengths) are (a) Csaltσ3m = 0.020 (κ1 = 2.0), (b) Csaltσ3m = 0.12 (κ1 = 0.8), (c) Csaltσ3m = 0.88 (κ1 = 0.3), and (d) Csaltσ3m = 2.0 (κ1 = 0.2).

A closer analysis shows that DNA expansion at a higher salt concentration corresponds to a discontinuous unfolding transition with the unwrapping of a DNA chain around protamine molecules (complex d in part B of Figure 5). Because this unwrapping is caused by weakening of the DNAprotamine electrostatic attraction, this DNA expansion at a high salt concentration is a common phenomenon regardless of the value of Q*. However, the DNA expansion at a lower salt concentration is induced for different reasons. In contrast to the case with a high

ARTICLE

salt concentration, the electrostatic attraction between DNA and protamine molecules is enhanced, and thus, the wrapping of DNA around individual protamines becomes tighter with a decrease in the salt concentration. Along with this tighter wrapping, the whole complex becomes more compact provided that the bare charge of DNA is sufficiently compensated by bound protamine molecules, that is, Q/zmNm , 1 so that the complex is close to the electrostatic neutral point, as in the case of Np = 20. However, if the complex as a whole has a substantial net charge Q*, the effect of the resultant electrostatic self-energy becomes progressively important and the complex becomes elongated with a decrease in the salt concentration (complexes a and b in part B of Figure 5). These features are in good agreement with the experimental results. Note that the reentrant behavior of the DNA size as a function of the monovalent-salt concentration is a rather general characteristic of a system composed of DNA and strong condensing agents with relatively a large number of positive charges, where Q* = Q(sys) holds due to strong electrostatic binding. A similar reentrant conformational change has long been known to exist for chromatin, where histone proteins act as a strong condensing agent. In the case of weaker condensing agents such as spermine, an excess amount of condensing agents must be added to induce compaction: |Q(sys)|.0. Only a fraction of the added condensing agents are involved in DNA compaction to achieve Q* ≈ 0, whereas the rest are free from it and constitute a particle reservoir. In this case, a decrease in the salt concentration leads to ion-exchange between the monovalent salt and z-valent condensing agents, which makes the compact state more stable. Thus, the elongated state at a low salt concentration is a unique feature seen in the system with strong condensing agents. A comment may be useful regarding the correspondence between a real system and our simplified model. As we stated above, we have adopted a beadspring model to represent a DNA chain for which the model parameters are selected so as to realize a semiflexible nature (σm,lp, L) and strong electrostatic attraction between the DNA and protamine molecules. We have also assumed that protamine molecules are spherical to extract generic electrostatic features. In addition, we have not included the salt concentration explicitly, but adopted the DebyeH€uckel approximation to express its effect.32 To compare the results of the experiment and the numerical simulation, we can set σm = 2 nm, where the diameter of DNA σm is a unit length in the model. Regardless of the above simplification, our model can reproduce the characteristic behavior of Rg depending on the salt concentration around a salt concentration of Csaltσ3m ≈ 1S[Csalt] ≈ 2.0 104 M (Figure 5). This indicates that our model is valid for a qualitative understanding of the experimental phenomena.

’ CONCLUSIONS DNA compaction and packaging in sperm nuclei is a very important phenomenon. Several studies have suggested that the poor packaging of sperm chromatin might be correlated with an increase in DNA damage. We recently demonstrated that the poor packing of DNA by protamine increases its radiosensitivity.33 In the present study, we demonstrated that the most efficient DNA compaction (collapse of a single DNA molecule) by protamine is achieved at physiological salt concentrations, and also offered theoretical considerations. This finding provides additional information on the role of protamine in living cells. 4458



’ AUTHOR INFORMATION Corresponding Author

*Tel/Fax: þ81 77 561 5064, e-mail: [email protected].

ARTICLE

(32) Zinchenko, A. A.; Sakaue, T.; Araki, S.; Yoshikawa, K.; Baigl, D. J. Phys. Chem. B 2007, 111, 3019–3031. (33) Suzuki, M.; Crozatier, C.; Yoshikawa, K.; Mori, T.; Yoshikawa, Y. Chem. Phys. Lett. 2009, 480, 113–117.

’ ACKNOWLEDGMENT This work was supported in part by Grants-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan (18GS0421) and from Japan Society for the Promotion of Science (22510123). ’ REFERENCES (1) Hancock, R. Semin. Cell Dev. Biol. 2007, 18, 668–675. (2) Bloomfield, V. A. Curr. Opin. Struct. Biol. 1996, 6, 334–341. (3) Schiessel, H. J. Phys.: Condens. Matter 2003, 15, R699–R774. (4) Gosule, L. C.; Schellman, J. A. Nature 1976, 259, 333–335. (5) Takahashi, M.; Yoshikawa, K.; Vasilevskaya, V. V.; Khokhlov, A. R. J. Phys. Chem. B 1997, 101, 9396–9401. (6) Raspaud, E.; Olvera de la Crus, M.; Sikorav, J.-L. Biophys. J. 1998, 74, 381–393. (7) Nguyen, T. T.; Rouzina, I.; Shklovskii, B. I. J. Chem. Phys. 2000, 112, 2562–2568. (8) Murayama, Y.; Sakamaki, Y.; Sano, M. Phys. Rev. Lett. 2003, 90, 018102. (9) Yang, J.; Rau, D. C. Biophys. J. 2005, 89, 1932–1940. (10) Widom, J.; Baldwin, R. L. J. Mol. Biol. 1980, 144, 431–453. (11) Yamasaki, Y.; Yoshikawa, K. J. Am. Chem. Soc. 1997, 119, 10573–10578. (12) Lerman, L. S. Proc. Natl. Acad. Sci. U. S. A. 1971, 68, 1886–1890. (13) Vasilevskaya, V. V.; Khokhlov, A. R.; Matsuzawa, Y.; Yoshikawa, K. J. Chem. Phys. 1995, 102, 6595–6602. (14) Mel’nikov, S. M.; Sergeyev, V. G.; Yoshikawa, K. J. Am. Chem. Soc. 1995, 117, 2401–2408. (15) Yoshihara, C.; Shew, C.-Y.; Ito, T.; Koyama, Y. Biophys. J. 2010, 98, 1257–1266. (16) Allen, M. J.; Bradbury, E. M.; Balhorn, R. Nucl. Acids. Res. 1997, 25, 2221–2226. (17) Raspaud, E.; Pelta, J.; de Frutos, M.; Livolant, F. Phys. Rev. Lett. 2006, 97, 068103. (18) Besteman, K.; van Eljk, K.; Lemay, S. G. Nature Phys. 2007, 3, 641–644. (19) Noguchi, H.; Saito, S.; Kidoaki, S.; Yoshikawa, K. Chem. Phys. Lett. 1996, 261, 527–533. (20) Tsumoto, K.; Luckel, F.; Yoshikawa, K. Biophys. Chem. 2003, 106, 23–29. (21) Brewer, L. R.; Corzett, M.; Balhorn, R. Science 1999, 286, 120–123. (22) Hud, N. V.; Allen, M. J.; Downing, K. H.; Lee, J.; Balhorn, R. Biochem. Biophys. Res. Commun. 1993, 193, 1347–1354. (23) Balhorn, R.; Brewer, L.; Corzett, M. Mol. Reprod. Dev. 2000, 56, 230–234. (24) Vilfan, I. D.; Conwell, C. C.; Hud, N. V. J. Biol. Chem. 2004, 279, 20088–20095. (25) Pinto, M. F. V.; Moran, M. C.; Miguel, M. G.; Lindman, B.; Jurado, A. S.; Pais, A. A. C. C. Biomacromolecules 2009, 10, 1319–1323. (26) Porschke, D. J. Mol. Biol. 1991, 222, 423–433. (27) Toma, A. C.; de Frutos, M.; Livolant, F.; Raspaud, E. Biomacromolecules 2009, 10, 2129–2134. (28) Gunther, K.; Mertig, M.; Seidel, R. Nucleic Acids Res. 2010, 38, 6526–6532. (29) Haaf, T.; Ward, D. C. Exp. Cell. Res. 1995, 219, 604–611. (30) Zlatanova, J.; Leuba, S. H.; van Holde, K. Biophys. J. 1998, 74, 2554–2566. (31) Yoshikawa, Y.; Velichko, Y. S.; Ichiba, Y.; Yoshikawa, K. Eur. J. Biochem. 2001, 268, 2593–2599. 4459


Salt Has a Biphasic Effect on the Higher-Order Structure of a DNA

Recommend Documents