|
1. |
Anomalous size dependence of the non-linear mobility of DNA |
|
PhysChemComm,
Volume 3,
Issue 11,
2000,
Page 61-63
Leonid L. Frumin,
Preview
|
|
摘要:
Anomalous size dependence of the non-linear mobility of DNA Leonid L. Frumin,† Sergey E. Peltek,§ Shmuel Bukshpan, Victor V. Chasovskikh† and Gleb V. Zilberstein* Proteologics (Israel) Ltd, Rabin Science Park, Lev HaTamar Bld., 4 Pekeris Street, Rehovot, 76702, Israel. E-mail: gleb_z@netvision.net.il Received 25th September 2000, Accepted 11th October 2000, Published 16th October 2000 A strong drift velocity non-linearity of the (doublestranded) DNA molecules vs. the electric field was discovered in gel electrophoresis. The non-linear drift velocity of the longer molecules is higher than that of the short ones, depending on the molecular size in a complex fashion. This behavior contrasts with that of the ordinary, linear, drift velocity where short molecules move faster than the long ones.The molecular size dependence of the non-linear velocity or the non-linear mobility has a nonmonotonous wave-like character. The non-linear mobility offers possibilities of manipulating the drift velocity at will—the DNA fragments of different size can be made to move in opposite directions in pulsed field gel electrophoresis. Recently, some remarkable DNA properties were discovered in the electric fields,1–3 relevant for the physics of macromolecules and biopolymers. Interestingly, these new phenomena in the DNA behavior in the electric fields may also be observed within the common gel electrophoresis and pulsed field gel electrophoresis techniques.4,5 Gel electrophoresis is one of the most frequently used separation techniques for biological macromolecules.The molecules migrate in support medium (gel), and are separated according to their mobility, determined by their charge and the gel composition, when an external electric field is applied. In comparatively weak fields, the molecular drift velocity, Ud, is proportional to the external electric field E: Ud = bE, the molecular mobility b being virtually field-independent. The macromolecules, existing initially in the form of a stochastic coil, start to deform in higher electric fields, aligning gradually along the external electric field vector. This makes the molecular mobility dependent on the electric field. McDonnell et al.6 were among the first to report such a dependence. The non-linear molecular mobility of DNA in gels had been discussed in the most detailed way by Lumpkin et al.7 They concluded that in moderate fields the non-linear mobility term b is proportional to the field amplitude squared: b = mE2, being independent of the molecule length.The respective non-linear velocity term is proportional to the field cubed: Ud = mE3. More complex dependence of the DNA molecules mobility on an electric field predicts the revised biased reptation model including fluctuation (BRF).8 The present paper presents an experimental study of the DNA non-linear mobility in moderate electric fields, and of its dependence upon the macromolecule length. To observe the non-linear term, we used an asymmetric periodic electric field with a zero average value, like in zero field electrophoresis.9 For instance, we applied a field of 10 V cm–1 for 10 s in one direction, and then a field of 3.33 V cm–1 for 30 s in the opposite direction.The average field value is obviously zero, while the average field cubed value DOI: 10.1039/b007756i is not. In the course of 23 h, after many periods of the electric field, we observed a noticeable movement of the DNA molecules, caused exclusively by the non-linear velocity term. We have even managed to separate the molecules as a function of their non-linear mobility. Fig. 1 presents the results of such non-linear electrophoresis (NEP) of the DNA molecules from the HIND III DNA hydrolysate of phage Lambda. Three lanes with several fractions each are visible in Fig.1. Note that the leading fraction at lanes 1 and 2, containing the molecules of the largest DNA fragment of 23.1 kbp, is Fig. 1 Phage Lambda/DNA HIND III fragments of 23.1 kbp and 4.4 kbp (lane 1); 23.1 kbp, 6.6 kbp and 2.3 kbp (lane 2); 9.4 kbp and 2.3 kbp (lane 3), were separated by the NEP at 10 V cm–1 (10 s) and –3.33 V cm–1 (33 s) in 0.75% agarose gel. Agarose (Ultra pure DNA Grade from BioRad Laboratories, Richmond, CA) was dissolved by boiling in tris-acetate buffer (89 mM tris base, 89 mM acetic acid and 2 mM ethylenediaminetetraacetic acid—EDTA). After cooling to 60 °C, 40 ml of agarose solutions at 0.75% w/v concentration was poured into 10 × 10 cm2 gel vessel, where plastic start slot formers were placed, designated by 's'. The gel slab height was 4 mm.The solidified gel was submerged in tris-acetate EDTA buffer, with a depth of 3 mm above the upper gel surface. The lambda/DNA HIND III fragments, dissolved in a loading buffer, were introduced into the start slot. Electrophoresis was performed for 48 h in a horizontal gel apparatus. The separated DNA fragments were stained with ethydium bromide, and visualized with an ultraviolet transilluminator (Model BioDoc II/NT, Biometra analytical GmbH, Germany). The voltage at the gel ends was controlled using special platinum metering electrodes of 0.2 mm in diameter. A computer-controlled voltage source was used to maintain the potential difference. Control was accomplished using a LAB PC+ unit from National Instruments. PhysChemComm, 2000, 11the most prominent and moves well ahead of all the other fractions.This is the most important difference between the non-linear electrophoresis and the stationary electrophoresis, where the smallest fragments move the fastest. This experiment yielded a notable result: the zero average field has not only caused molecular movement, but also separated the DNA fractions as a function of their nonlinear mobility. We can notice that a similar method of experiment was suggested by Turmel et al. and is known as zero-integrated field electrophoresis (ZIFE).13 But there is no term ‘nonlinear mobility’ in this work. We use regimes that are similar to ZIFE, and we understand that when <E> = 0 it is possible to get effects that relate with non-linear mobility of DNA.That is why the most important thing that we use is a new parameter for DNA molecules manipulations in variable fields. In order to compare the NEP with the conventional constant field electrophoresis, and to clarify the sequence of the DNA fragments in the NEP, we performed a twodimensional electrophoresis. In this experiment, the DNA molecules were first subject to the already described NEP procedure. Secondly, adding identical DNA marker fragments in an additional slot, a steady-state EP was performed in an orthogonal direction. At the second stage the markers, for which the sequence is well known, should arrive at the same positions as the identical fractions obtained in the NEP experiment. Fig.2 presents the results of such a two-dimensional electrophoresis. The results presented in Fig. 2 demonstrate, that the NEP produces the fraction order opposite to that of the conventional EP. Thus, the heavier fragments have a larger non-linear mobility and move ahead of the light fractions. However, we find band inversion comparing fraction 2 (9.4 kbp) to fraction 3 (6.6 kbp). Fig. 2 2D-electrophoresis map: 38 h of the non-linear electrophoresis (NEP) in 0.75% agarose gel: 7 V cm–1 × 5 s, –2.33 V cm–1 × 15 s in one direction, and 24 h of steady-state EP at 1.5 V cm–1 in an orthogonal (vertical) direction, compared with a conventional steady-state EP of Lambda/DNA HIND III marker (the vertical track on the left).The fragments are numbered as follows: 23.1 kbp (1), 9.4 kbp (2), 6.6 kbp (3), 4.4 kbp (4), 2.5 kbp (5), 2.3 kbp (6). We assumed at first that this non-monotonous dependence of the non-linear mobility on the DNA molecule size is probably related to the specific conformation of the 9.4 kbp DNA fragment. The anomalous mobility of this fragment is equally manifest in the conventional EP. This phenomenon was related9,10 to the non-monotonous dependence of the mobility on the molecule length, due to formation of stable compact loop-like conformations, for certain proportions between the average pore size and the molecule length. However, a study of the phenomenon has shown that the fraction inversion is in no way specific to the lambda/DNA HIND III marker.In particular, the 2.5 kbp molecular ruler marker from BioRad (CN170-8205) has also revealed a similar non-monotonous wave-like dependence of the molecular mobility upon the linear size of the DNA molecule. This ruler has 15 DNA bands spaced from 2.5 to 35 kbp. Fig. 3 shows the result of the 2D electrophoresis of this marker, represented in the same coordinates of Fig. 2. The wave-like structure of the size dependence of the non-linear mobility contains several zones. The non-linear mobility grows with the size between 2 and 4 kbp, subsequently falling from 4 to 6 kbp. After 6 kbp, the non-linear mobility again grows, getting to a maximum at 22–25 kbp, falling again for the molecules larger than 25 kbp.The wave-like dependence of non-linear mobility from size is similar to the well-known phenomena ‘band-inversion’14 in the other variants of electrophoresis with alternating fields (field inversion gel electrophoresis—FIGE or CHEF). But all the works concerning ‘band inversion’ describe only the behavior of linear mobility of DNA molecules. Also we used periods of electric signal much more longer than 'anti-resonance' ones that are used in Fig. 3 2D-electrophoresis 2.5 kbp BioRad Laboratories ruler map. 23 h NEP in 0.75% agarose gel: 7 V cm–1 × 5 s, –2.33 V cm–1 × 15 s in one direction, and 24 h of steady-state EP at 2.33 V cm–1 in an orthogonal (vertical) direction, compared to the conventional steady-state EP of the DNA 2.5 kbp ruler (the vertical track on the right).Fig.4 Fraction drift direction control by non-linear mobility. Phage lambda/DNA HIND III fragments of 23.1 and 4.4 kbp (upper lane); 23.1, 6.6 and 2.3 kbp (medial lane); 9.4 and 2.3 kbp (lower lane), were separated by the NEP at 10 V cm–1 × 20 s and –2.3 V cm–1 × 100 s in 0.75% agarose gel. The linear drift velocity component is positive, while the non-linear—negative. The symbol 'S' correspond to start slots. FIGE. It is important because FIGE uses in most of the cases minimum mobility of DNA molecules. This is an important difference in the methods of our experiments. An effort was made to explain the wave-like behavior of DNA fractions in NEP as compared to conventional EP in the framework of the biased reptation model7,10,11 (BRM) and also biased reptation model including fluctuations (BRF).8 The BRM model predicts a weak monotonous growth of the non-linear mobility and does not explain its fine wavelike structure.Also BRF does not contain non-monotonic non-linear mobility term. As follows from the results of the 2D EP presented in Figs. 2 and 3, the dependence of the non-linear mobility on the molecular size has several extremes. Thus, it possibly may not be described by BRM and also BRF models. Another remarkable feature of the NEP is that it permits to manipulate the DNA fragment drift velocity. For instance, the variable field parameters of the NEP experiment may be chosen such that the linear and the non-linear components of the average drift velocity would have opposite signs for the different bands, which will drift in opposite directions.Fig. 4 presents an example of such an experiment. The light fractions have high linear mobility and thus move in the direction contrary to that of the heavy fractions, having a relatively high non-linear mobility. Note that the field switching times used varied from 1 to 100 s, always remaining much higher than those characteristic for the minimal mobility effect.12 This experiment proves that the non-linear component of drift velocity of molecules can not be featured by an even degree of a field. Only an odd degree for the dependence of drift velocity (even degree for mobility) can result in the opposite bias of fractions from a place of start.Overall, we conclude that the non-linear electrophoresis provides an important addition to the existing set of molecular biology techniques. Possible correlation of the molecular conformational structure with their non-linear mobility clearly requires further experimental and theoretic studies, being of major importance for the modern physics of macromolecules. We would like to thank professors P.G. Righetti, I.V. Khmelinskii and U. Halavi for valuable advice and useful discussions. References 1 J. Han and H. G. Craighead, Science, 2000, 208, 1026. 2 J. Han, S. W. Turner and H. G. Craighead, Phys. Rev. Lett., 1999, 83, 1688. 3 T. A. J. Duke and R.H. Austin, Phys. Rev. Lett., 1994, 72, 2117. 4 D. C. Schwartz and C.R. Cantor, Cell, 1984, 37, 67. 5 G. F. Carle, M. Frank and M.W. Olson, Nucleic Acids Res., 1984, 12, 5647. 6 U. W. McDonnell, M. N. Simon and F. W. Studier, J. Mol. Biol., 1977, 110, 119. 7 O. J. Lumpkin, P. Dejardin and B. Zimm, Biopolymers, 1985, 24, 1573. 8 T. Duke, J.-L. Viovy and N. Semenov, Biopolymers, 1994, 34, 239. 9 J. Noolandi and C. Turmel, in Methods in Molecular Biology, ed. M. Bunneister and L. Ulanovsky, The Humana Press, Totowa, 1992, ch. 7, p. 73. 10 J. Noolandi, J. Rousseau, G. W. Slater and M. Lalande, Phys. Rev. Lett., 1988, 58, 2428. 11 G. W. Slater, C. Turmel, M. Lalande and J. Noolandi, Biopolymers, 1989, 28, 1793. 12 T. A. J. Duke, Phys. Rev. Lett., 1989, 62, 2877. 13 C. Turmel, E. Brassard, G. W. Slater and J. Noolandi, Nucleic Acids Res., 1990, 18, 569. 14 C. Heller and F. M. Pohl, Nucleic Acids Res., 1991, 18, 6299. † Present address: Novosibirsk State University, 2 Pirogova Street, Novosibirsk-90, 630090, Russia. § Present Address: Institute of Cytology and Genetics, 4 Koptiug Prospect, Novosibirsk-90, 630090, Russia. PhysChemComm © The Royal Society of Chemistry 2000
ISSN:1460-2733
DOI:10.1039/b007756i
出版商:RSC
年代:2000
数据来源: RSC
|
|