- Email: [email protected]
DNA condensations on mica surfaces induced collaboratively by alcohol and hexammine cobalt
Colloids and Surfaces B: Biointerfaces 83 (2011) 61–68
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
DNA condensations on mica surfaces induced collaboratively by alcohol and hexammine cobalt Yanwei Wang a,b , Shiyong Ran b , Baoyuan Man a , Guangcan Yang b,∗ a b
School of Physics and Electronic Sciences, Shandong Normal University, Jinan 250014, China School of Physics and Electronic Information, Wenzhou University, Wenzhou 325035, China
a r t i c l e
i n f o
Article history: Received 19 July 2010 Received in revised form 25 October 2010 Accepted 25 October 2010 Available online 3 November 2010 Keywords: DNA condensation Alcohol Hexammine cobalt Electrostatic interaction Toroids
a b s t r a c t We performed systematic studies of -DNA condensation on mica surfaces induced by alcohol and hexammine cobalt (III) [Co(NH3 )6 3+ ] using atomic force microscopy (AFM). The critical condensation concentration for [Co(NH3 )6 3+ ] was found to be about 10 M; the DNA molecules extended freely on mica when the concentration was below the critical value. The morphology of condensed DNA became more compact with increasing concentration. At about 500 M [Co(NH3 )6 3+ ] concentration, no condensation patterns could be observed due to charge inversion of the compact structures resulting in failure of adhesion to the positively charged surfaces. The critical concentration for alcohol was about 15% (v/v). At this concentration, a few intramolecular loops could be observed in the AFM images. With increasing ethanol concentration the condensation pattern became more complicated ranging from flower-like to pancake-like. When the solution contained both alcohol and hexammine cobalt (III), DNA condensation patterns could be observed even when the concentrations of the two condensation agents were lower than their critical values. We observed this phenomenon by adding mixtures of 10% alcohol and 8 M hexammine cobalt (III) to DNA solutions. The condensation patterns were more compact than those of the condensation agents separately. Typical toroids were found at an appropriate alcohol and hexammine cobalt (III) concentration. The collaborative condensation phenomenon was analyzed by electrostatic interaction and charge neutralization. © 2010 Elsevier B.V. All rights reserved.
1. Introduction DNA condensation is the collapse of extended DNA chains into compact, ordered particles containing only one or a few molecules. It can be induced by the addition of condensing agents and/or ligands (e.g., multivalent cations of valence three or greater, cationic polypeptides such as polylysine, basic proteins, alcohols, and neutral crowding polymers) [1]. Condensation and compactation of DNA have attracted in a large amount of interest due to its importance both in technological and biomedical applications, particularly in recent years as potential vehicles for gene delivery and gene transfection [2–4]. By adding condensing agent to very dilute DNA solution at low ionic strength, condensed toroids and rods can be observed [5]. With the increased DNA concentrations in solutions, some liquid crystal structures are also formed [6,7]. Given the high negative surface charge, a strong repulsive energy barrier is expected against compaction, surprisingly thus, DNA in solution spontaneously undergoes a transition from an extended random coil conformation to a condensed state upon addition of condens-
∗ Corresponding author. Tel.: +86 577 86689033. E-mail addresses: [email protected], [email protected] (G. Yang). 0927-7765/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.colsurfb.2010.10.040
ing agents. DNA condensation may proceed as a collapse of single DNA molecules or as intermolecular aggregation. The condensation mechanism by multivalent cations is likely a neutralization of the large negative charge of the DNA phosphates, thereby overcoming the columbic repulsive forces to produce the compact conformation. For neutral condensing agents like alcohols, the condensation mechanism is related to the solvent properties. In a good solvent, the monomers effectively repel each other, preferring to be surrounded by a solvent. This effect leads to a swollen coil conformation for flexible polymers in a good solvent. In a poor solvent, conversely, the monomers try to exclude the solvent and effectively attract one another. As a result, flexible chains form a compact globule of roughly spherical shape to minimize the contacts between monomers and solvent. The collapsed state configurations and the pathways to their formation are the results of the interplay between two opposing forces: the bending force related to the chain stiffness and the attractive force due to the poor solvency of the environment. When the two different kinds of condensing agents exist in solution, it is possible that a mixed mechanism drives the process of condensation. Multivalent cations may cause localized bending or distortion of the DNA, which may facilitate condensation. Over the past several years, theorists have proposed attractive mechanisms that cause
62
Y. Wang et al. / Colloids and Surfaces B: Biointerfaces 83 (2011) 61–68
Fig. 1. Morphologires of condensed -DNA at 15% (A), 20% (B) and 30% (C-1 and C-2) ethanol. A very wide range of structures is seen at different ethanol concentration. The A, B and C-1 figures are rinsed with water solution and air-drying. The C-2 figure is not rinsed with water and subsequently dried under ambient conditions.
DNA to condense in the presence of multivalent cations [8,9]. The Manning’s counterion condensing theory [10] explained that the DNA condensing occurs in the presence of multivalent cations when ∼90% of the negative charge of its backbone is neutralized [11]. Polyamines spermidine, spermine and multivalent cation hexammine cobalt (III) [Co(NH3 )6 3+ ] are the most common cations used as condensing agents. Among of them, Co(NH3 )6 3+ is a more potent condensing agent than spermidine as it produces more tightly packed particles, and induces a high degree of bending [12,13]. Alcohols and neutral or anionic polymers can also provoke DNA condensation. High concentrations of ethanol are commonly used to precipitate DNA, but only under carefully controlled conditions, it can produce particles of well-defined morphologies [14,15]. Atomic force microscopy (AFM) experiments have examined the effect of alcohol on the structures of DNA confined to mica in the presence of Mg2+ [16]. At high alcohol concentration (>20%), various higher order structures are formed, including flowers and toroids. Besides, the DNA molecules adsorbed onto mica can be subsequently condensed by a brief rinse with anhydrous alcohol [17,18]. The majority of these surface-directed and ethanol-directed condensed structures are toroids, but a small fraction of rods can also be observed [19]. Both alcohol and multivalent cations can separately cause transitions in DNA helical structure. It might be anticipated that the two agents would work together cooperatively [20]. Although 80% ethanol is commonly used to precipitate DNA, as little as 15–20% ethanol will also induce condensation if Co(NH3 )6 3+ is also added to a solution at low ionic strength. In this paper, we systematically investigate the -DNA condensations on mica surface induced by alcohol and hexammine cobalt (III) [Co(NH3 )6 3+ ] with atomic force microscopy (AFM). The independent critical condensation concentrations of the two condensing agents are measured carefully. It is found that when both alcohol and Co(NH3 )6 3+ coexist in the solution, DNA condensation patterns can be observed even when the concentration of the two condensation agents are lower than their critical values. The collaborative phenomena can be interpreted by electrostatic interaction and solvent effects.
a conductivity less than 1 × 10−6 −1 cm−1 . Mica was cut into approximately 1 cm2 square pieces and mica surfaces were always freshly cleaved before use. Hexammine cobalt (Co(NH3 )6 3+ ) was purchased as chloride salts and dissolved in pure water before use. Other chemicals were all purchased from Sigma–Aldrich. 2.2. Sample preparation. The protocols for sample preparation are as follows: Method 1: The stock solution of 500 ng/l -DNA was diluted with 1× TE containing MgCl2 . Experiments were performed by mixing ethanol solution of appropriate volume concentration to DNA solution. The final concentration of DNA is 2.5 ng/l. The final concentration of MgCl2 is 5 mM. Samples were incubated at room temperature for different time. A 20 l liquid of the mixture was deposited onto freshly cleaved mica and incubated for 5 min. Following incubation, the mica surface was washed with 100 Milli-Q filtered water for 5 times (∼5 s) to remove the free and excess molecules, and then rapidly blown dried using a gentle stream of nitrogen gas. As an alternative, the mica surface was not rinsed with water and subsequently dried under ambient conditions. Method 2: The stock solution of 500 ng/l -DNA was diluted to a final concentration of 5 ng/l with 1× TE containing 10 mM MgCl2 . Co(NH3 )6 3+ solution of different concentration was mixed with an equal volume of DNA solutions. The mixture was incubated for 1 h at 25 ◦ C. Ten microliters of this solution was dropped immediately onto freshly cleaved mica surface. After air-drying for 3 min, the mica was gently rinsed with pure water in the same way as described as above, then dried in air, and left sealed. Method 3: The mixture solution of Co(NH3 )6 3+ and alcohol were added to an equal volume of 5 ng/l DNA solutions containing 10 mM MgCl2 . The solution was incubated for 1 h at 25 ◦ C. Ten microliters of this solution was deposited immediately onto freshly cleaved mica surface. After air-drying for 3 min, the mica was dealt with in the same way as described as above. 2.3. Atomic force microscopy measurements
2. Experimental procedures 2.1. Materials. Bacteria -phage DNA was purchased from New England Biolabs and used without further purification. As received from the manufacturer, the -phage DNA stock solution had a concentration of 500 ng/L. The solvent is 1×TE buffer, which is composed of 10 mM Tris–HCl, pH 8.0, and 1 mM EDTA. Anhydrous ethanol and methanol were purchased from Sinopharm Chemical Reagent, Beijing in China. Water was deionized and purified by a Millipore system and had
The imaging was performed in air with a multi-mode AFM with nanoscope controller (SPM-9600, Shimadzu, Kyoto, Japan) in the tapping-mode. All AFM images shown are height images with scan speeds of ∼2 Hz and data collection at 512 × 512 pixels. The length, height, and width of the DNA in AFM imaging were measured manually using the off-line analysis software with AFM. All the experiments were repeated at least three times to ensure consistent results. The experimental error in critical concentration for hexammine cobalt (III) measurement is ±1 M and the experimental error in critical concentration for ethanol measurement is ±3% (v/v). Typical data was presented in this paper.
Y. Wang et al. / Colloids and Surfaces B: Biointerfaces 83 (2011) 61–68
63
3. Results and discussion 3.1. The DNA condensation induced by ethanol In the first protocol, DNA solutions were incubated with different concentrations of ethanol, followed by a rinse with water solution and then rapidly blown dried using a gentle stream of nitrogen gas as shown in Fig. 1A, B and C-1. Alternatively, the specimens are not rinsed with water and subsequently dried under ambient conditions as shown in Fig. 1C-2. The critical concentration for alcohol induced DNA condensation is about 15% (v/v). When the alcohol concentration in DNA solutions decreased to 15%, there are few intermolecular contacts, but individual molecules have a few intramolecular loops. At ethanol >15%, multi-molecular complexes appear, including those with flower-like morphologies. As the ethanol concentration is increased to 30%, a range of more complex flowers structures appears as in Fig. 1C-1. A typical flower pattern has a core in the centre and DNA wraps around it at high density. There is a general tendency for the structures to become more complex with increasing ethanol concentration. Ethanol can alter DNA conformation from B-form to A- or C-form, depending on the nature of the counterion [21–23]. It has been shown that 80% ethanol is required for a complete B-to-A transition of a DNA in the presence of cations such as Na+ , K+ or Cs+ [22]. In contrast, the presence of the strongly hydrated counterion Mg2+ is found to prevent the formation of the A-form, as ethanol concentration increases [23]. Divalent metal cations do not provoke condensation in water at room temperatures, but an opposite effect may be observed at somewhat elevated temperatures or in water–alcohol mixtures. MgCl2 can also induce DNA condensation in alcohol–water mixtures [24,25]. The rinse with water was performed without excess Mg2+ or free DNA to prevent the formation of salt crystals or condensation induced by magnesium cations on the mica surface. In Fig. 1C-2, the AFM imaging reveals morphologies which are not flowers. However, the morphologies of rods are seen on mica surface. The surface of mica has to be rinsed with water, so that there are no salt crystals or other condensation dots on it. Certainly, DNA–DNA attractions due to alcohol exclusion would be very easily reversed with a water rinse. In our experiments, as the time of depositing on mica is long enough, the interaction between the DNA and mica is strong. As a result, the decondensation on mica surface does not occur clearly. At low ethanol concentration, the condensation morphologies of flowers could be expected. On the other hand, at high concentration of ethanol with carefully controlled conditions particles of well defined morphology can be obtained, and can be used to precipitate DNA. As the ethanol concentration is increased to >80% in our experiments, a range of more complex pancakes and snowflake structures appear, as shown in Fig. 2. There the morphologies are replaced by more extensively aggregated structures. It is well known that high concentrations of ethanol affect the solubility and conformation of DNA. 3.2. The DNA condensation induced by hexammine cobalt In the second condensation protocol, the stock solution of 500 ng/l -DNA was diluted to a final concentration of 5 ng/l with 1× TE containing 10 mM MgCl2 . Co(NH3 )6 3+ at different concentration was mixed with an equal volume of DNA solutions. The final concentrations of Co(NH3 )6 3+ are 10 M, 20 M, 40 M, 60 M, 100 M, 200 M, 300 M and 500 M. A striking feature of DNA condensation is that the dimensions and morphology of condensed DNA particles are largely independent of the size of the DNA. DNA fragments shorter than about 400 base pairs will not condense into orderly, discrete particles. For example, 150 bp mononucleosomal DNA cannot be condensed from dilute solution into discrete
Fig. 2. Range of intermediates of condensed -DNA at 80% ethanol and the time of incubation is about 10 h. A range of more complex pancakes and snowflake structures appears.
particles of orderly morphology [26,27]. At least several hundred base pairs must interact, either intramolecularly or intermolecularly, to form a stably condensed particle. We use the -DNA which has 48,502 bp in our experiment serving in favor of condensation. Samples prepared by pre-mixing DNA and Co(NH3 )6 3+ in tube are imaged and presented in Fig. 3. Fig. 3A–H represents the morphologies of DNA on mica surfaces with the Co(NH3 )6 3+ concentrations of 10, 20, 40, 60, 100, 200, 300 and 500 M, respectively. When the concentration of Co(NH3 )6 3+ is less than 10 M, no condensation is observed and DNA is freely extended on mica, consistent with the existing results [28]. However, when it crosses over the critical concentration, the condensation grows gradually. The condensation emerges very explicitly at 20 M as we can see in Fig. 3B. When the concentration of the condensing agent increases from 20 to 200 M, we can see similar patterns but with increasing clustering as shown in Fig. 3C–F. In these cases, DNA segments wrap around the core to form a flat disk, or in some circumstances, wrap in a way like the petals around the core of a flower such as in Fig. 3C. Most of these condensates are single layered, con-
64
Y. Wang et al. / Colloids and Surfaces B: Biointerfaces 83 (2011) 61–68
Fig. 3. The condensed -DNA in hexammine cobalt at different concentration. A–H shows respectively the condensation patterns under 10, 20, 40, 60, 100, 200, 300 and 500 M hexammine cobalt.
Fig. 4. (A) The condensed -DNA in hexammine cobalt at concentration of 20 M. The structures like toroids and flowers could be observed on the mica surfaces. (B). It is a magnification of toroid of (A). Scale bar represent 100 nm. (C) Cross-sectional profile of the toroid in (B) along the marked line.
Y. Wang et al. / Colloids and Surfaces B: Biointerfaces 83 (2011) 61–68
65
Fig. 5. The condensed -DNA by adding mixtures of alcohols and Co(NH3 )6 3+ to DNA solutions. (A)–(C) Shows the condensation morphologies adding 10% (A), 20% (B), 30% (C) ethanol concentration to the condensed -DNA in 10 M hexammine cobalt.
taining more than one DNA molecules, but still rather flat. When Co(NH3 )6 3+ concentration increases to 300 M, the flower patterns disappear and other patterns like flat pancakes emerge. Therefore, multivalent cations not only profoundly affect the structure of DNA, both by condensing it and by modifying its local structure, but also influence their adsorption to surface. When the Co(NH3 )6 3+ concentration increases further to 500 M, the results become remarkably different with hardly any pattern found (Fig. 3H). We tried to image the condensates at much higher concentration up to 5 mM, and the similar results were obtained. Condensation of DNA by multivalent cations generally forms particles of regular shape, rather than the randomly entangled aggregates that might be naively anticipated. In fact, coiled DNA formed in the solution may be attached by Co(NH3 )6 3+ onto mica surfaces and form flat patterns. On the other hand, for the compact morphologies formed in the solution, it is difficult to get adhered to the charged surfaces, and to sustain the rinsing and nitrogen blowing. The structures observed here reflect a compromise between the effects of condensing agents and DNA interactions with mica. When DNA binds to mica, the Co3+ /Mg2+ ions on it must be released. The energy lost in releasing ions increases with increasing salt concentration. At higher Co3+ concentrations, mica binding tends to lost since the energy gained in mica is less than the energy lost in releasing Co3+ /Mg2+ . On the other hand, the flower-like structures are observed rather than the more compact toroids or globules because of release of Co3+ that mediates attraction to gain stronger DNA–mica interactions [29]. In our experiments, the most common structures observed are the flat patterns of DNA, while the regular structures like toroids could hardly be observed on the mica surfaces. We still found some regular structures as typically shown in Fig. 4A, where a DNA toroid coexists with a neighboring flower pattern. The toroidal confor-
mation of the condensed DNA, with a well-defined central hole surrounded by circumferentially wound double helical strands, is one of the most striking aspects of the condensation phenomenon. Fig. 4B is a typical toroid structure of -DNA induced by 20 M hexammine cobalt. It is measured by the online analyzing software with the AFM and the result is shown in Fig. 4C. The outer diameter of the toroid in Fig. 4B is about 146.56 nm while the inner diameter is about 44.16 nm. We measured all the toroids obtained and their dimensions range from 120 nm to 200 nm, which is consistent with the results obtained by electron microscopy [20]. Since DNA is a semiflexible polymer molecule and the attractive forces between DNA segments are rather weak, the DNA molecules in poor solute tend to form compact toroidal structure in solution due to the equilibrium between the exclusion and bending energy when the concentration of condensing agent is low. They can be observed less often in AFM imaging than electron microscopy, because the compact toroids are difficult to adsorb to mica surface after withstanding of water rinsing and nitrogen blowing during sample preparation. 3.3. The DNA condensation induced by the mixture of alcohol and hexammine cobalt In the third condensation protocol, the stock solution of 500 ng/L -DNA was diluted to a final concentration of 5 ng/L with 1× TE containing 10 mM MgCl2 . We explored the condensation by adding mixtures of different concentrations of alcohols and Co(NH3 )6 3+ to an equal volume of DNA solutions. The condensation patterns are more compact than those by the corresponding condensation agents separately. For a specific condensing agent, condensation occurs only above the critical concentration for given DNA and salt concentrations.
Fig. 6. The condensed -DNA by adding mixtures of alcohols and Co(NH3 )6 3+ to DNA solutions. (A)–(C) Shows the condensation morphologies adding 10% (A), 20% (B), 30% (C) methanol concentration to the condensed -DNA in 10 M hexammine cobalt.
66
Y. Wang et al. / Colloids and Surfaces B: Biointerfaces 83 (2011) 61–68
Fig. 7. The condensed -DNA by adding mixtures of alcohols and Co(NH3 )6 3+ to DNA solutions. (A) Shows the condensation morphologies adding 5% ethanol to the condensed -DNA in 5 M hexammine cobalt. (B) Shows the condensation morphologies adding 10% ethanol to the condensed -DNA in 8 M hexammine cobalt.
In our experiment, we measured the critical concentration of Co(NH3 )6 3+ and ethanol, which are 10 M and 15% (v/v), respectively. The extent of condensation near criticality is a linear function of excess [Co(NH3 )6 3+ ]. When the concentration of a condensing agent is decreased to below its critical value, the condensation of DNA cannot be observed in AFM imaging. When more than one condensing agents exist in solution, behavior of DNA condensation needs further investigation. Some results by dynamic light scattering and electron microscopy can be found in Ref. [20]. We explore in this direction by using Co(NH3 )6 3+ and alcohol with AFM. We attempted to use ethanol and methanol, respectively and obtained the similar results. The samples were prepared by adding mixtures of 10%, 20%, 30% alcohols and 10 M Co(NH3 )6 3+ to DNA solutions and incubating for an hour, where hexamine cobalt is at the critical value. The resultant AFM images are shown in Figs. 5 and 6, which correspond to ethanol and methanol respectively. We can compare the cases by 20% ethanol, 10 M hexamine cobalt and their combination, which are shown in Figs. 1B, 3A and 5B. It can
Fig. 8. (A) The condensed -DNA in hexammine cobalt (100 M) and 30% ethanol solution. (B) The toriodal morphologies of -DNA in hexammine cobalt (100 M) and 5% ethanol solution.
be seen that the condensation patterns become much more apparent and more compact than those induced by the corresponding condensing agents separately. We may explain the observation by considering the electrostatic interaction. The addition of alcohol lowers the dielectric constants of the solution. The low dielectric solute (alcohol in the case) may also have lower affinity for bimolecular surfaces, when compared to water. The rate as well as the extent of condensation increases as the dielectric constant is lowered. We observed the similar phenomena for ethanol and methanol. Theoretically, ethanol may induce stronger condensation than methanol because of a lower dielectric constant (25 vs. 30). But the resultant difference is hard to distinguish in the corresponding AFM images (Figs. 5 and 6). Based on the observation, we can conclude that alcohols like methanol and ethanol enhance DNA condensation via changing the dielectric force. Interestingly, when both alcohol and Co(NH3 )6 3+ exist in the DNA solution, DNA condensation patterns can appear even when both the concentrations are lower than their critical values, shown in Fig. 7. In Fig. 7A, we can see an initial stage of condensation of
Y. Wang et al. / Colloids and Surfaces B: Biointerfaces 83 (2011) 61–68
67
Fig. 9. Structures of DNA condensation induced by 10% alcohol and 15 M Co(NH3 )6 3+ .The comparisons of the AFM imagines rinsed with water for two times, four times, ten times and twenty times.
DNA by adding mixtures of 5% alcohols and 5 M Co(NH3 )6 3+ to DNA solutions. These concentrations are much lower than their critical values. In Fig. 7B the condensation becomes very apparent and quite compact in the core, while the concentrations of alcohol (10%) and Co(NH3 )6 3+ (8 M) are still below their critical ones. This is an interesting collaborative phenomenon. We may ascribe the observation to the electrostatic interaction in solution. Alcohols are poor solutes for DNA solution, and are typically excluded from the DNA phase, which could be said to “crowd” the solution, so as to exert a force on it and force DNA together and finally push DNA toward more compact states [30]. We consider that the observation is a collaborative phenomena rather than an additive one. As we can see in Fig. 1B (20% ethanol) and Fig. 3A (10 M hexammine cobalt), where the condensation is quite loose so as hard to judge. However when these two are mixed together, a much more apparent and compact condensation is obtained, e.g. in Fig. 5B. Changes in the morphologies of DNA complex in the presence of ethanol may well indicate the crucial role of the electrostatic force in causing DNA condensation. The electrostatic interactions become stronger as the dielectric constant ε decreases. Lowering ε lowers the effective phosphate charge by increasing counterion condensation. This should facilitate DNA condensation. The cooperation between the two agents amplifies the condensing effect significantly. This cooperation is worth to investigate further theoretically in the future. At an appropriate ethanol and Co(NH3 )6 3+ concentration, we can see that the DNA condenses into three-dimensional toroids on surfaces, which are shown in Fig. 8. In our experiment, the toroids are more easily formed at a lower ethanol concentration since the condensation undergoes at a lower rate. At high ethanol concentration, the compact DNA condensates induced by Co(NH3 )6 3+ form so rapidly that it is difficult for bound cations to have time to diffuse along the duplex backbone to their favorable binding sites to form a regular structure. In sample preparation, water rinse may induce the decondensation of DNA and influence its observation on mica surface. Specifically, DNA–DNA attractions due to alcohol exclusion may be reversed with a water rinse. We tested this water rinse effect, as shown in Fig. 9. In the study of mixed condensating agents (10% alcohol and 15 M Co(NH3 )6 3+ ), we made the comparisons of the AFM images rinsed with water for different duration, which are shown in Fig. 8. We can see that there is nearly no change of the condensation structures. It seems that the interaction between the DNA and mica is strong enough to sustain the water rinse because of Co(NH3 )6 3+ binding to DNA, which is different from alcohol exclusion. 4. Conclusions Compaction of DNA by condensing agents can provide insights into DNA assembly processes, which relate closely to the essence of gene transfection and gene therapy. We have studied system-
atically the DNA condensations confined to a mica surface in the presence of varying concentrations of hexammine cobalt and alcohols, independently. The critical condensation concentration for Co(NH3 )6 3+ is about 10 M, while for ethanol it is 15% (v/v). The DNA molecules extend freely on mica when the concentration is below the critical values. The morphologies of DNA condensation become more compact with the increasing concentration. When both alcohol and Co(NH3 )6 3+ exist in the solution, DNA condensation patterns can be observed even when the concentration of each of the two condensation agents are lower than their own individual critical values. Addition of hexammine cobalt and alcohol leads to form flat and single-layered DNA patterns on mica surfaces. Their shapes are nearly circular with a core in the centre, and the DNA strands wrap around the core in a compact and relatively ordered manner. Through the comparison of DNA–hexammine cobalt condensation and DNA–alcohol condensation, we have found that the condensation of DNA in three-dimensional toroids on surfaces is difficult because of the rather strong attractive forces between DNA segments. It is difficult for bound cations to have sufficient time to diffuse along the duplex backbone to their favorable binding sites. Due to such restrictions on forming 3D structures, DNA wraps to form a rather flat and compact patterns, like flowers, pancakes and snowflakes. When a mixture of alcohol and Co(NH3 )6 3+ is used, the addition of alcohol tends to produce more condensed structures when the condensation is provoked by [Co(NH3 )6 3+ ] resulting a much stronger condensation. Acknowledgements This work is partially supported by the National Key Basic Research Project of China (Grant No. 2007CB935900), the National Natural Science Foundation of China (Grant No. 10974146 and 20934004) and Zhejiang Provincial Natural Science Foundation (Grant No. Y6090222). References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16]
V.A. Bloomfield, Curr. Opin. Struct. Biol. 6 (1996) 334. N.V. Hud, I.D. Vilfan, Annu. Rev. Biophys. Biomol. Struct. 34 (2005) 295–318. J. Pelta, D. Durand, J. Doucet, F. Livolant, Biophys. J. 71 (1996) 48–63. E. Raspaud, D. Durand, F. Livolant, Biophys. J. 88 (2005) 392–403. D.D. Lasic, Lopsomes in Gene Delivery, CRC Press, Boca Raton, FL, 1997. N.S. Templeton, D.D. Lasic, Mol. Biotechnol. 11 (1999) 175–183. L. Huang, M.C. Hung, E. Wagner, Non Viral Vecters for Gene Delivery, Academic Press, New York, 1999. W. Gelbart, R. Bruinsma, P. Pincus, V. Parsegian, Phys. Today 53 (2000) 38. H. Strey, R. Podgornik, D. Rau, V. Parsegian, Curr. Opin. Struct. Biol. 8 (1998) 309. G. Manning, J. Chem. Phys. 51 (1969) 924. R. Wilson, V. Bloomfield, Biochemistry 18 (1979) 2192. J. Schellman, N. Parthasarathy, J. Mol. Biol. 175 (1984) 313. J. Widom, R. Baldwin, J. Mol. Biol. 144 (1980) 431. T.H. Eickbush, E.N. Moudrianakis, Cell 13 (1976) 295. D. Lang, J. Mol. Biol. 78 (1973) 247. Y. Fang, T. Spisz, J. Hoh, Nucl. Acids Res. 27 (1999) 1943.
68
Y. Wang et al. / Colloids and Surfaces B: Biointerfaces 83 (2011) 61–68
[17] Z. Xiao, M. Xu, K. Sagisaka, D. Fujita, Thin Solid films 438–439 (2003) 114–117. [18] Y. Song, Z. Li, Z. Liu, G. Wei, L. Wang, L. Sun, C. Guo, Y. Sun, T. Yang, J. Phys. Chem. B 110 (2006) 10792. [19] C. Zhang, J. van der Maarel, J. Phys. Chem. B-Condens. Phase 112 (2008) 3552. [20] P. Arscott, C. Ma, J. Wenner, V. Bloomfield, Biopolymers 36 (1995) 345. [21] V.I. Ivanov, D. Krylov, Methods Enzymol. 211 (1992) 111. [22] A. Rupprecht, J. Piskur, G. Lahajnar, Biopolymers 34 (1994) 897. [23] J. Schultz, A. Rupprecht, Z. Song, G. Lahajnar, Biophys. J. 66 (1994) 810.
[24] R.W. Wilson, V.A. Bloomfild, Biochemistry 18 (1979) 2192. [25] H. Votavova, D. Kucerova, J. Felsberg, J. Sponar, J. Biomol. Struct. Dynam. 4 (1986) 477. [26] V.A. Bloomfiled, Biopolymers 31 (1991) 1471. [27] J. Widom, R.L. Baldwin, J. Mol. Biol. 144 (1980) 431. [28] S. He, P.G. Arscott, V.A. Bloomfield, Biopolymers 53 (2000) 329. [29] The anonymous referee provides this nice explanation. [30] A. Hultgren, D.C. Rau, Biochemistry 43 (2004) 8272.
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
DNA condensations on mica surfaces induced collaboratively by alcohol and hexammine cobalt Yanwei Wang a,b , Shiyong Ran b , Baoyuan Man a , Guangcan Yang b,∗ a b
School of Physics and Electronic Sciences, Shandong Normal University, Jinan 250014, China School of Physics and Electronic Information, Wenzhou University, Wenzhou 325035, China
a r t i c l e
i n f o
Article history: Received 19 July 2010 Received in revised form 25 October 2010 Accepted 25 October 2010 Available online 3 November 2010 Keywords: DNA condensation Alcohol Hexammine cobalt Electrostatic interaction Toroids
a b s t r a c t We performed systematic studies of -DNA condensation on mica surfaces induced by alcohol and hexammine cobalt (III) [Co(NH3 )6 3+ ] using atomic force microscopy (AFM). The critical condensation concentration for [Co(NH3 )6 3+ ] was found to be about 10 M; the DNA molecules extended freely on mica when the concentration was below the critical value. The morphology of condensed DNA became more compact with increasing concentration. At about 500 M [Co(NH3 )6 3+ ] concentration, no condensation patterns could be observed due to charge inversion of the compact structures resulting in failure of adhesion to the positively charged surfaces. The critical concentration for alcohol was about 15% (v/v). At this concentration, a few intramolecular loops could be observed in the AFM images. With increasing ethanol concentration the condensation pattern became more complicated ranging from flower-like to pancake-like. When the solution contained both alcohol and hexammine cobalt (III), DNA condensation patterns could be observed even when the concentrations of the two condensation agents were lower than their critical values. We observed this phenomenon by adding mixtures of 10% alcohol and 8 M hexammine cobalt (III) to DNA solutions. The condensation patterns were more compact than those of the condensation agents separately. Typical toroids were found at an appropriate alcohol and hexammine cobalt (III) concentration. The collaborative condensation phenomenon was analyzed by electrostatic interaction and charge neutralization. © 2010 Elsevier B.V. All rights reserved.
1. Introduction DNA condensation is the collapse of extended DNA chains into compact, ordered particles containing only one or a few molecules. It can be induced by the addition of condensing agents and/or ligands (e.g., multivalent cations of valence three or greater, cationic polypeptides such as polylysine, basic proteins, alcohols, and neutral crowding polymers) [1]. Condensation and compactation of DNA have attracted in a large amount of interest due to its importance both in technological and biomedical applications, particularly in recent years as potential vehicles for gene delivery and gene transfection [2–4]. By adding condensing agent to very dilute DNA solution at low ionic strength, condensed toroids and rods can be observed [5]. With the increased DNA concentrations in solutions, some liquid crystal structures are also formed [6,7]. Given the high negative surface charge, a strong repulsive energy barrier is expected against compaction, surprisingly thus, DNA in solution spontaneously undergoes a transition from an extended random coil conformation to a condensed state upon addition of condens-
∗ Corresponding author. Tel.: +86 577 86689033. E-mail addresses: [email protected], [email protected] (G. Yang). 0927-7765/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.colsurfb.2010.10.040
ing agents. DNA condensation may proceed as a collapse of single DNA molecules or as intermolecular aggregation. The condensation mechanism by multivalent cations is likely a neutralization of the large negative charge of the DNA phosphates, thereby overcoming the columbic repulsive forces to produce the compact conformation. For neutral condensing agents like alcohols, the condensation mechanism is related to the solvent properties. In a good solvent, the monomers effectively repel each other, preferring to be surrounded by a solvent. This effect leads to a swollen coil conformation for flexible polymers in a good solvent. In a poor solvent, conversely, the monomers try to exclude the solvent and effectively attract one another. As a result, flexible chains form a compact globule of roughly spherical shape to minimize the contacts between monomers and solvent. The collapsed state configurations and the pathways to their formation are the results of the interplay between two opposing forces: the bending force related to the chain stiffness and the attractive force due to the poor solvency of the environment. When the two different kinds of condensing agents exist in solution, it is possible that a mixed mechanism drives the process of condensation. Multivalent cations may cause localized bending or distortion of the DNA, which may facilitate condensation. Over the past several years, theorists have proposed attractive mechanisms that cause
62
Y. Wang et al. / Colloids and Surfaces B: Biointerfaces 83 (2011) 61–68
Fig. 1. Morphologires of condensed -DNA at 15% (A), 20% (B) and 30% (C-1 and C-2) ethanol. A very wide range of structures is seen at different ethanol concentration. The A, B and C-1 figures are rinsed with water solution and air-drying. The C-2 figure is not rinsed with water and subsequently dried under ambient conditions.
DNA to condense in the presence of multivalent cations [8,9]. The Manning’s counterion condensing theory [10] explained that the DNA condensing occurs in the presence of multivalent cations when ∼90% of the negative charge of its backbone is neutralized [11]. Polyamines spermidine, spermine and multivalent cation hexammine cobalt (III) [Co(NH3 )6 3+ ] are the most common cations used as condensing agents. Among of them, Co(NH3 )6 3+ is a more potent condensing agent than spermidine as it produces more tightly packed particles, and induces a high degree of bending [12,13]. Alcohols and neutral or anionic polymers can also provoke DNA condensation. High concentrations of ethanol are commonly used to precipitate DNA, but only under carefully controlled conditions, it can produce particles of well-defined morphologies [14,15]. Atomic force microscopy (AFM) experiments have examined the effect of alcohol on the structures of DNA confined to mica in the presence of Mg2+ [16]. At high alcohol concentration (>20%), various higher order structures are formed, including flowers and toroids. Besides, the DNA molecules adsorbed onto mica can be subsequently condensed by a brief rinse with anhydrous alcohol [17,18]. The majority of these surface-directed and ethanol-directed condensed structures are toroids, but a small fraction of rods can also be observed [19]. Both alcohol and multivalent cations can separately cause transitions in DNA helical structure. It might be anticipated that the two agents would work together cooperatively [20]. Although 80% ethanol is commonly used to precipitate DNA, as little as 15–20% ethanol will also induce condensation if Co(NH3 )6 3+ is also added to a solution at low ionic strength. In this paper, we systematically investigate the -DNA condensations on mica surface induced by alcohol and hexammine cobalt (III) [Co(NH3 )6 3+ ] with atomic force microscopy (AFM). The independent critical condensation concentrations of the two condensing agents are measured carefully. It is found that when both alcohol and Co(NH3 )6 3+ coexist in the solution, DNA condensation patterns can be observed even when the concentration of the two condensation agents are lower than their critical values. The collaborative phenomena can be interpreted by electrostatic interaction and solvent effects.
a conductivity less than 1 × 10−6 −1 cm−1 . Mica was cut into approximately 1 cm2 square pieces and mica surfaces were always freshly cleaved before use. Hexammine cobalt (Co(NH3 )6 3+ ) was purchased as chloride salts and dissolved in pure water before use. Other chemicals were all purchased from Sigma–Aldrich. 2.2. Sample preparation. The protocols for sample preparation are as follows: Method 1: The stock solution of 500 ng/l -DNA was diluted with 1× TE containing MgCl2 . Experiments were performed by mixing ethanol solution of appropriate volume concentration to DNA solution. The final concentration of DNA is 2.5 ng/l. The final concentration of MgCl2 is 5 mM. Samples were incubated at room temperature for different time. A 20 l liquid of the mixture was deposited onto freshly cleaved mica and incubated for 5 min. Following incubation, the mica surface was washed with 100 Milli-Q filtered water for 5 times (∼5 s) to remove the free and excess molecules, and then rapidly blown dried using a gentle stream of nitrogen gas. As an alternative, the mica surface was not rinsed with water and subsequently dried under ambient conditions. Method 2: The stock solution of 500 ng/l -DNA was diluted to a final concentration of 5 ng/l with 1× TE containing 10 mM MgCl2 . Co(NH3 )6 3+ solution of different concentration was mixed with an equal volume of DNA solutions. The mixture was incubated for 1 h at 25 ◦ C. Ten microliters of this solution was dropped immediately onto freshly cleaved mica surface. After air-drying for 3 min, the mica was gently rinsed with pure water in the same way as described as above, then dried in air, and left sealed. Method 3: The mixture solution of Co(NH3 )6 3+ and alcohol were added to an equal volume of 5 ng/l DNA solutions containing 10 mM MgCl2 . The solution was incubated for 1 h at 25 ◦ C. Ten microliters of this solution was deposited immediately onto freshly cleaved mica surface. After air-drying for 3 min, the mica was dealt with in the same way as described as above. 2.3. Atomic force microscopy measurements
2. Experimental procedures 2.1. Materials. Bacteria -phage DNA was purchased from New England Biolabs and used without further purification. As received from the manufacturer, the -phage DNA stock solution had a concentration of 500 ng/L. The solvent is 1×TE buffer, which is composed of 10 mM Tris–HCl, pH 8.0, and 1 mM EDTA. Anhydrous ethanol and methanol were purchased from Sinopharm Chemical Reagent, Beijing in China. Water was deionized and purified by a Millipore system and had
The imaging was performed in air with a multi-mode AFM with nanoscope controller (SPM-9600, Shimadzu, Kyoto, Japan) in the tapping-mode. All AFM images shown are height images with scan speeds of ∼2 Hz and data collection at 512 × 512 pixels. The length, height, and width of the DNA in AFM imaging were measured manually using the off-line analysis software with AFM. All the experiments were repeated at least three times to ensure consistent results. The experimental error in critical concentration for hexammine cobalt (III) measurement is ±1 M and the experimental error in critical concentration for ethanol measurement is ±3% (v/v). Typical data was presented in this paper.
Y. Wang et al. / Colloids and Surfaces B: Biointerfaces 83 (2011) 61–68
63
3. Results and discussion 3.1. The DNA condensation induced by ethanol In the first protocol, DNA solutions were incubated with different concentrations of ethanol, followed by a rinse with water solution and then rapidly blown dried using a gentle stream of nitrogen gas as shown in Fig. 1A, B and C-1. Alternatively, the specimens are not rinsed with water and subsequently dried under ambient conditions as shown in Fig. 1C-2. The critical concentration for alcohol induced DNA condensation is about 15% (v/v). When the alcohol concentration in DNA solutions decreased to 15%, there are few intermolecular contacts, but individual molecules have a few intramolecular loops. At ethanol >15%, multi-molecular complexes appear, including those with flower-like morphologies. As the ethanol concentration is increased to 30%, a range of more complex flowers structures appears as in Fig. 1C-1. A typical flower pattern has a core in the centre and DNA wraps around it at high density. There is a general tendency for the structures to become more complex with increasing ethanol concentration. Ethanol can alter DNA conformation from B-form to A- or C-form, depending on the nature of the counterion [21–23]. It has been shown that 80% ethanol is required for a complete B-to-A transition of a DNA in the presence of cations such as Na+ , K+ or Cs+ [22]. In contrast, the presence of the strongly hydrated counterion Mg2+ is found to prevent the formation of the A-form, as ethanol concentration increases [23]. Divalent metal cations do not provoke condensation in water at room temperatures, but an opposite effect may be observed at somewhat elevated temperatures or in water–alcohol mixtures. MgCl2 can also induce DNA condensation in alcohol–water mixtures [24,25]. The rinse with water was performed without excess Mg2+ or free DNA to prevent the formation of salt crystals or condensation induced by magnesium cations on the mica surface. In Fig. 1C-2, the AFM imaging reveals morphologies which are not flowers. However, the morphologies of rods are seen on mica surface. The surface of mica has to be rinsed with water, so that there are no salt crystals or other condensation dots on it. Certainly, DNA–DNA attractions due to alcohol exclusion would be very easily reversed with a water rinse. In our experiments, as the time of depositing on mica is long enough, the interaction between the DNA and mica is strong. As a result, the decondensation on mica surface does not occur clearly. At low ethanol concentration, the condensation morphologies of flowers could be expected. On the other hand, at high concentration of ethanol with carefully controlled conditions particles of well defined morphology can be obtained, and can be used to precipitate DNA. As the ethanol concentration is increased to >80% in our experiments, a range of more complex pancakes and snowflake structures appear, as shown in Fig. 2. There the morphologies are replaced by more extensively aggregated structures. It is well known that high concentrations of ethanol affect the solubility and conformation of DNA. 3.2. The DNA condensation induced by hexammine cobalt In the second condensation protocol, the stock solution of 500 ng/l -DNA was diluted to a final concentration of 5 ng/l with 1× TE containing 10 mM MgCl2 . Co(NH3 )6 3+ at different concentration was mixed with an equal volume of DNA solutions. The final concentrations of Co(NH3 )6 3+ are 10 M, 20 M, 40 M, 60 M, 100 M, 200 M, 300 M and 500 M. A striking feature of DNA condensation is that the dimensions and morphology of condensed DNA particles are largely independent of the size of the DNA. DNA fragments shorter than about 400 base pairs will not condense into orderly, discrete particles. For example, 150 bp mononucleosomal DNA cannot be condensed from dilute solution into discrete
Fig. 2. Range of intermediates of condensed -DNA at 80% ethanol and the time of incubation is about 10 h. A range of more complex pancakes and snowflake structures appears.
particles of orderly morphology [26,27]. At least several hundred base pairs must interact, either intramolecularly or intermolecularly, to form a stably condensed particle. We use the -DNA which has 48,502 bp in our experiment serving in favor of condensation. Samples prepared by pre-mixing DNA and Co(NH3 )6 3+ in tube are imaged and presented in Fig. 3. Fig. 3A–H represents the morphologies of DNA on mica surfaces with the Co(NH3 )6 3+ concentrations of 10, 20, 40, 60, 100, 200, 300 and 500 M, respectively. When the concentration of Co(NH3 )6 3+ is less than 10 M, no condensation is observed and DNA is freely extended on mica, consistent with the existing results [28]. However, when it crosses over the critical concentration, the condensation grows gradually. The condensation emerges very explicitly at 20 M as we can see in Fig. 3B. When the concentration of the condensing agent increases from 20 to 200 M, we can see similar patterns but with increasing clustering as shown in Fig. 3C–F. In these cases, DNA segments wrap around the core to form a flat disk, or in some circumstances, wrap in a way like the petals around the core of a flower such as in Fig. 3C. Most of these condensates are single layered, con-
64
Y. Wang et al. / Colloids and Surfaces B: Biointerfaces 83 (2011) 61–68
Fig. 3. The condensed -DNA in hexammine cobalt at different concentration. A–H shows respectively the condensation patterns under 10, 20, 40, 60, 100, 200, 300 and 500 M hexammine cobalt.
Fig. 4. (A) The condensed -DNA in hexammine cobalt at concentration of 20 M. The structures like toroids and flowers could be observed on the mica surfaces. (B). It is a magnification of toroid of (A). Scale bar represent 100 nm. (C) Cross-sectional profile of the toroid in (B) along the marked line.
Y. Wang et al. / Colloids and Surfaces B: Biointerfaces 83 (2011) 61–68
65
Fig. 5. The condensed -DNA by adding mixtures of alcohols and Co(NH3 )6 3+ to DNA solutions. (A)–(C) Shows the condensation morphologies adding 10% (A), 20% (B), 30% (C) ethanol concentration to the condensed -DNA in 10 M hexammine cobalt.
taining more than one DNA molecules, but still rather flat. When Co(NH3 )6 3+ concentration increases to 300 M, the flower patterns disappear and other patterns like flat pancakes emerge. Therefore, multivalent cations not only profoundly affect the structure of DNA, both by condensing it and by modifying its local structure, but also influence their adsorption to surface. When the Co(NH3 )6 3+ concentration increases further to 500 M, the results become remarkably different with hardly any pattern found (Fig. 3H). We tried to image the condensates at much higher concentration up to 5 mM, and the similar results were obtained. Condensation of DNA by multivalent cations generally forms particles of regular shape, rather than the randomly entangled aggregates that might be naively anticipated. In fact, coiled DNA formed in the solution may be attached by Co(NH3 )6 3+ onto mica surfaces and form flat patterns. On the other hand, for the compact morphologies formed in the solution, it is difficult to get adhered to the charged surfaces, and to sustain the rinsing and nitrogen blowing. The structures observed here reflect a compromise between the effects of condensing agents and DNA interactions with mica. When DNA binds to mica, the Co3+ /Mg2+ ions on it must be released. The energy lost in releasing ions increases with increasing salt concentration. At higher Co3+ concentrations, mica binding tends to lost since the energy gained in mica is less than the energy lost in releasing Co3+ /Mg2+ . On the other hand, the flower-like structures are observed rather than the more compact toroids or globules because of release of Co3+ that mediates attraction to gain stronger DNA–mica interactions [29]. In our experiments, the most common structures observed are the flat patterns of DNA, while the regular structures like toroids could hardly be observed on the mica surfaces. We still found some regular structures as typically shown in Fig. 4A, where a DNA toroid coexists with a neighboring flower pattern. The toroidal confor-
mation of the condensed DNA, with a well-defined central hole surrounded by circumferentially wound double helical strands, is one of the most striking aspects of the condensation phenomenon. Fig. 4B is a typical toroid structure of -DNA induced by 20 M hexammine cobalt. It is measured by the online analyzing software with the AFM and the result is shown in Fig. 4C. The outer diameter of the toroid in Fig. 4B is about 146.56 nm while the inner diameter is about 44.16 nm. We measured all the toroids obtained and their dimensions range from 120 nm to 200 nm, which is consistent with the results obtained by electron microscopy [20]. Since DNA is a semiflexible polymer molecule and the attractive forces between DNA segments are rather weak, the DNA molecules in poor solute tend to form compact toroidal structure in solution due to the equilibrium between the exclusion and bending energy when the concentration of condensing agent is low. They can be observed less often in AFM imaging than electron microscopy, because the compact toroids are difficult to adsorb to mica surface after withstanding of water rinsing and nitrogen blowing during sample preparation. 3.3. The DNA condensation induced by the mixture of alcohol and hexammine cobalt In the third condensation protocol, the stock solution of 500 ng/L -DNA was diluted to a final concentration of 5 ng/L with 1× TE containing 10 mM MgCl2 . We explored the condensation by adding mixtures of different concentrations of alcohols and Co(NH3 )6 3+ to an equal volume of DNA solutions. The condensation patterns are more compact than those by the corresponding condensation agents separately. For a specific condensing agent, condensation occurs only above the critical concentration for given DNA and salt concentrations.
Fig. 6. The condensed -DNA by adding mixtures of alcohols and Co(NH3 )6 3+ to DNA solutions. (A)–(C) Shows the condensation morphologies adding 10% (A), 20% (B), 30% (C) methanol concentration to the condensed -DNA in 10 M hexammine cobalt.
66
Y. Wang et al. / Colloids and Surfaces B: Biointerfaces 83 (2011) 61–68
Fig. 7. The condensed -DNA by adding mixtures of alcohols and Co(NH3 )6 3+ to DNA solutions. (A) Shows the condensation morphologies adding 5% ethanol to the condensed -DNA in 5 M hexammine cobalt. (B) Shows the condensation morphologies adding 10% ethanol to the condensed -DNA in 8 M hexammine cobalt.
In our experiment, we measured the critical concentration of Co(NH3 )6 3+ and ethanol, which are 10 M and 15% (v/v), respectively. The extent of condensation near criticality is a linear function of excess [Co(NH3 )6 3+ ]. When the concentration of a condensing agent is decreased to below its critical value, the condensation of DNA cannot be observed in AFM imaging. When more than one condensing agents exist in solution, behavior of DNA condensation needs further investigation. Some results by dynamic light scattering and electron microscopy can be found in Ref. [20]. We explore in this direction by using Co(NH3 )6 3+ and alcohol with AFM. We attempted to use ethanol and methanol, respectively and obtained the similar results. The samples were prepared by adding mixtures of 10%, 20%, 30% alcohols and 10 M Co(NH3 )6 3+ to DNA solutions and incubating for an hour, where hexamine cobalt is at the critical value. The resultant AFM images are shown in Figs. 5 and 6, which correspond to ethanol and methanol respectively. We can compare the cases by 20% ethanol, 10 M hexamine cobalt and their combination, which are shown in Figs. 1B, 3A and 5B. It can
Fig. 8. (A) The condensed -DNA in hexammine cobalt (100 M) and 30% ethanol solution. (B) The toriodal morphologies of -DNA in hexammine cobalt (100 M) and 5% ethanol solution.
be seen that the condensation patterns become much more apparent and more compact than those induced by the corresponding condensing agents separately. We may explain the observation by considering the electrostatic interaction. The addition of alcohol lowers the dielectric constants of the solution. The low dielectric solute (alcohol in the case) may also have lower affinity for bimolecular surfaces, when compared to water. The rate as well as the extent of condensation increases as the dielectric constant is lowered. We observed the similar phenomena for ethanol and methanol. Theoretically, ethanol may induce stronger condensation than methanol because of a lower dielectric constant (25 vs. 30). But the resultant difference is hard to distinguish in the corresponding AFM images (Figs. 5 and 6). Based on the observation, we can conclude that alcohols like methanol and ethanol enhance DNA condensation via changing the dielectric force. Interestingly, when both alcohol and Co(NH3 )6 3+ exist in the DNA solution, DNA condensation patterns can appear even when both the concentrations are lower than their critical values, shown in Fig. 7. In Fig. 7A, we can see an initial stage of condensation of
Y. Wang et al. / Colloids and Surfaces B: Biointerfaces 83 (2011) 61–68
67
Fig. 9. Structures of DNA condensation induced by 10% alcohol and 15 M Co(NH3 )6 3+ .The comparisons of the AFM imagines rinsed with water for two times, four times, ten times and twenty times.
DNA by adding mixtures of 5% alcohols and 5 M Co(NH3 )6 3+ to DNA solutions. These concentrations are much lower than their critical values. In Fig. 7B the condensation becomes very apparent and quite compact in the core, while the concentrations of alcohol (10%) and Co(NH3 )6 3+ (8 M) are still below their critical ones. This is an interesting collaborative phenomenon. We may ascribe the observation to the electrostatic interaction in solution. Alcohols are poor solutes for DNA solution, and are typically excluded from the DNA phase, which could be said to “crowd” the solution, so as to exert a force on it and force DNA together and finally push DNA toward more compact states [30]. We consider that the observation is a collaborative phenomena rather than an additive one. As we can see in Fig. 1B (20% ethanol) and Fig. 3A (10 M hexammine cobalt), where the condensation is quite loose so as hard to judge. However when these two are mixed together, a much more apparent and compact condensation is obtained, e.g. in Fig. 5B. Changes in the morphologies of DNA complex in the presence of ethanol may well indicate the crucial role of the electrostatic force in causing DNA condensation. The electrostatic interactions become stronger as the dielectric constant ε decreases. Lowering ε lowers the effective phosphate charge by increasing counterion condensation. This should facilitate DNA condensation. The cooperation between the two agents amplifies the condensing effect significantly. This cooperation is worth to investigate further theoretically in the future. At an appropriate ethanol and Co(NH3 )6 3+ concentration, we can see that the DNA condenses into three-dimensional toroids on surfaces, which are shown in Fig. 8. In our experiment, the toroids are more easily formed at a lower ethanol concentration since the condensation undergoes at a lower rate. At high ethanol concentration, the compact DNA condensates induced by Co(NH3 )6 3+ form so rapidly that it is difficult for bound cations to have time to diffuse along the duplex backbone to their favorable binding sites to form a regular structure. In sample preparation, water rinse may induce the decondensation of DNA and influence its observation on mica surface. Specifically, DNA–DNA attractions due to alcohol exclusion may be reversed with a water rinse. We tested this water rinse effect, as shown in Fig. 9. In the study of mixed condensating agents (10% alcohol and 15 M Co(NH3 )6 3+ ), we made the comparisons of the AFM images rinsed with water for different duration, which are shown in Fig. 8. We can see that there is nearly no change of the condensation structures. It seems that the interaction between the DNA and mica is strong enough to sustain the water rinse because of Co(NH3 )6 3+ binding to DNA, which is different from alcohol exclusion. 4. Conclusions Compaction of DNA by condensing agents can provide insights into DNA assembly processes, which relate closely to the essence of gene transfection and gene therapy. We have studied system-
atically the DNA condensations confined to a mica surface in the presence of varying concentrations of hexammine cobalt and alcohols, independently. The critical condensation concentration for Co(NH3 )6 3+ is about 10 M, while for ethanol it is 15% (v/v). The DNA molecules extend freely on mica when the concentration is below the critical values. The morphologies of DNA condensation become more compact with the increasing concentration. When both alcohol and Co(NH3 )6 3+ exist in the solution, DNA condensation patterns can be observed even when the concentration of each of the two condensation agents are lower than their own individual critical values. Addition of hexammine cobalt and alcohol leads to form flat and single-layered DNA patterns on mica surfaces. Their shapes are nearly circular with a core in the centre, and the DNA strands wrap around the core in a compact and relatively ordered manner. Through the comparison of DNA–hexammine cobalt condensation and DNA–alcohol condensation, we have found that the condensation of DNA in three-dimensional toroids on surfaces is difficult because of the rather strong attractive forces between DNA segments. It is difficult for bound cations to have sufficient time to diffuse along the duplex backbone to their favorable binding sites. Due to such restrictions on forming 3D structures, DNA wraps to form a rather flat and compact patterns, like flowers, pancakes and snowflakes. When a mixture of alcohol and Co(NH3 )6 3+ is used, the addition of alcohol tends to produce more condensed structures when the condensation is provoked by [Co(NH3 )6 3+ ] resulting a much stronger condensation. Acknowledgements This work is partially supported by the National Key Basic Research Project of China (Grant No. 2007CB935900), the National Natural Science Foundation of China (Grant No. 10974146 and 20934004) and Zhejiang Provincial Natural Science Foundation (Grant No. Y6090222). References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16]
V.A. Bloomfield, Curr. Opin. Struct. Biol. 6 (1996) 334. N.V. Hud, I.D. Vilfan, Annu. Rev. Biophys. Biomol. Struct. 34 (2005) 295–318. J. Pelta, D. Durand, J. Doucet, F. Livolant, Biophys. J. 71 (1996) 48–63. E. Raspaud, D. Durand, F. Livolant, Biophys. J. 88 (2005) 392–403. D.D. Lasic, Lopsomes in Gene Delivery, CRC Press, Boca Raton, FL, 1997. N.S. Templeton, D.D. Lasic, Mol. Biotechnol. 11 (1999) 175–183. L. Huang, M.C. Hung, E. Wagner, Non Viral Vecters for Gene Delivery, Academic Press, New York, 1999. W. Gelbart, R. Bruinsma, P. Pincus, V. Parsegian, Phys. Today 53 (2000) 38. H. Strey, R. Podgornik, D. Rau, V. Parsegian, Curr. Opin. Struct. Biol. 8 (1998) 309. G. Manning, J. Chem. Phys. 51 (1969) 924. R. Wilson, V. Bloomfield, Biochemistry 18 (1979) 2192. J. Schellman, N. Parthasarathy, J. Mol. Biol. 175 (1984) 313. J. Widom, R. Baldwin, J. Mol. Biol. 144 (1980) 431. T.H. Eickbush, E.N. Moudrianakis, Cell 13 (1976) 295. D. Lang, J. Mol. Biol. 78 (1973) 247. Y. Fang, T. Spisz, J. Hoh, Nucl. Acids Res. 27 (1999) 1943.
68
Y. Wang et al. / Colloids and Surfaces B: Biointerfaces 83 (2011) 61–68
[17] Z. Xiao, M. Xu, K. Sagisaka, D. Fujita, Thin Solid films 438–439 (2003) 114–117. [18] Y. Song, Z. Li, Z. Liu, G. Wei, L. Wang, L. Sun, C. Guo, Y. Sun, T. Yang, J. Phys. Chem. B 110 (2006) 10792. [19] C. Zhang, J. van der Maarel, J. Phys. Chem. B-Condens. Phase 112 (2008) 3552. [20] P. Arscott, C. Ma, J. Wenner, V. Bloomfield, Biopolymers 36 (1995) 345. [21] V.I. Ivanov, D. Krylov, Methods Enzymol. 211 (1992) 111. [22] A. Rupprecht, J. Piskur, G. Lahajnar, Biopolymers 34 (1994) 897. [23] J. Schultz, A. Rupprecht, Z. Song, G. Lahajnar, Biophys. J. 66 (1994) 810.
[24] R.W. Wilson, V.A. Bloomfild, Biochemistry 18 (1979) 2192. [25] H. Votavova, D. Kucerova, J. Felsberg, J. Sponar, J. Biomol. Struct. Dynam. 4 (1986) 477. [26] V.A. Bloomfiled, Biopolymers 31 (1991) 1471. [27] J. Widom, R.L. Baldwin, J. Mol. Biol. 144 (1980) 431. [28] S. He, P.G. Arscott, V.A. Bloomfield, Biopolymers 53 (2000) 329. [29] The anonymous referee provides this nice explanation. [30] A. Hultgren, D.C. Rau, Biochemistry 43 (2004) 8272.