ANALYTICAL SCIENCES NOVEMBER 2001, VOL. 17 2001 © The Japan Society for Analytical Chemistry
1253
Original Papers
Identifying the Orientation of DNA Fragment in Recombinant Plasmid by Capillary Electrophoresis with a Non-Gel Sieving Solution Bi-Feng LIU, Qi-Guang XIE, and Ying-Tang LU† Key Laboratory of MOE for Plant Developmental Biology, Wuhan University, Wuhan, 430072, P. R. China
It was demonstrated that a capillary electrophoresis (CE) method with a non-gel sieving solution has been developed to identify the orientation of DNA fragments in recombinant plasmids in molecular biology. The influences of the concentration of sieving polymer HEC, the applied electric field strength and sampling on CE separation were analyzed concerning the optimization of separation. YO-PRO-1 was used as a DNA intercalating reagent to facilitate fluorescence detection. Under the chosen conditions (buffer, 1 × TBE containing 1 µM YO-PRO-1 and 1.2% HEC; applied electric field strength, 200 V/cm; electrokinetic sampling: time, 5 s; voltage, –6 kV), three DNA markers (φ174/HaeIII, pBR322/HaeIII and λDNA/HindIII) were tested for further evaluating the relationship between the DNA size and the mobility. The established CE method conjugated with the enzymatic approach was successfully applied to identifying the DNA orientation of recombinant plasmid in transgene operations of a newly cloned gene from Arabidopsis Thaliana. (Received February 26, 2001; Accepted August 20, 2001)
Introduction Gene cloning is a basic operation in molecular biology.1 As a result, the analysis of DNA fragments has become routine work to monitor DNA manipulation in molecular cloning.2 Taking the transgene experiment as an example, a target DNA fragment is often cloned to a chosen plasmid that can self-replicate in cells, known as a building construct. Briefly, if the cyclic DNA of a plasmid is first cut into linear DNA by an endonuclease, and the target DNA fragment is ligated onto the linear plasmid end-to-end by DNA ligase, there would be two forms of recombinant plasmids with a differently orientated insert DNA fragment, theoretically with equal probability. The DNA orientation should be discriminated, because different orientations would have a quite divergently biological function when transferred into organisms. One would result in an overexpression of the gene, while the reversed one would lead to the transcription of antisense, which would in turn inhibit the expression. The most widely used method to identify the orientation of DNA is slab gel electrophoresis (SGE) combined with restriction enzyme digestion or DNA sequencing. Capillary electrophoresis (CE) is emerging as a novel technique in the area of bio-analysis, and has now been recognized as a powerful method to analyze biomacromolecules,3 such as nuclear acid, protein and saccharide, due to its highly resolving ability, low time and sample consumption and advanced automation. The analysis of DNA has been widely acknowledged. Many authors have provided excellent reviews4–10 concerning the practice of CE in such a field. In earlier days, a cross-linked polyacrylamide gel was employed as the sieving matrix to support the size-dependent † To whom correspondence should be addressed. E-mail:
[email protected]
separation mechanism, termed capillary gel electrophoresis (CGE),11,12 which could offer a resolution of a single base pair (bp) ranged from 15 bp to 500 bp. However, difficulties in preparing a capillary gel column, such as the formation of a bubble and the low duration of high voltage, have bottlenecked its further development. Recently, linear polymers have been popularly used in CE as the sieving matrix instead of the crosslinked gel, in terms of non-gel sieving or an entangled polymersolution CE,13,14 due to its facilitation15 of high separation speed, long read length and replaceable operation compared to CGE. Many polymers have proved to be useful, including linear polyacrylamide and its derivatives,14,16,17 cellulose and its derivatives,18,19 polyethene oxide20,21 and agarose22 etc. Currently, non-gel sieving CE is being extensively applied in DNA sequencing, characterizing the polymorphism of DNA, analyzing PCR products, gene diagnosis and therapy etc.3,23 In this study, a capillary zone electrophoresis method with non-gel sieving medium was successfully demonstrated to identify the orientation of DNA fragments in recombinant plasmids. The influences of some parameters on electrophoretic separation were briefly investigated. It showed a great prospect of CE as a convenient and mature technique for analyzing DNA in the biological field.
Experimental Materials Ethidium bromide (EB, Molecule Probe Inc., CA, USA) was prepared in water at a concentration of 10 µg ml–1. The concentration of a stock solution of YO-PRO-1 (Molecule Probe Inc., CA, USA) was 1 mM in DMSO. Hydroethylcellulose (HEC, 4000 cp in a 2% solution at 25˚C) was obtained from Sigma, USA. TBE buffer (89 mM Tris, 89 mM borate, 2 mM EDTA, pH 8.3), TE buffer (10 mM Tris–HCl,
1254
ANALYTICAL SCIENCES NOVEMBER 2001, VOL. 17
1 mM EDTA, pH 8.0), agarose gels were prepared as recommended by Molecular Clone. Water purified by a Millipore-Q system (Millipore, USA) and autoclaved by hightemperature (121˚C) and high-pressure (15 lbf) was used for the preparation of all solutions. All chemicals were of analytical grade. Carrier electrolyte (1 × TBE containing 1 µM YO-PRO1 and 0.4 – 2.0% HEC) for capillary electrophoresis was prepared daily, boiled in a water bath for 5 min and then centrifuged at 12000 rpm for 0.5 min prior to use. DNA standards: φX174/HaeIII standard containing 11 fragments was prepared in our lab by the digestion of φX174 plasmid with HaeIII endonuclease at a final concentration of 0.15 µg ml–1; pBR322/HaeIII standard containing 22 fragments at a concentration of 0.15 µg ml–1 and λDNA/HindIII standard containing 8 fragments at a concentration of 0.5 µg ml–1 were obtained from Zhongjian Bioengineer Inc., China. All of the standards were diluted by 20-fold with TE buffer for the injections of capillary electrophoresis.
µL 10 × buffer (60 mM Tris–HCl pH 7.5, 1 M NaCl, 60 mM MgCl2 and 10 mM DTT), on 1 µg plasmid DNA. Then, the products of digestion were subjected to CE.
Instrument Analyses were carried out on a home-made capillary electrophoresis system. It was based on an upright fluorescence microscope (Olympus BX60, Japan). A cooled charge-coupled device (cooled-CCD, Diagnostic Instrument Inc., USA) camera was employed as the detector. The high-voltage power supply was purchased from Shanghai Nuclear Research Institute, China. Home-coated fused-silica capillaries were used as the separation column. A 100 W high-pressure mercury lamp was used as the excitation radiation. The optical sub-system in the microscope consisted of a 20× objective (UPlanApo, NA 0.70), an NIB excitation cube including an excitation filter (e.c. 470 – 490 nm), a dichroic mirror (DM 510 nm) and a barrier filter (BA 515 nm). The main working preferences of the CCD camera are as follows: pixel depth, 8 bpp; gain, 1; binning, 4 × 4. Data were collected by a computer (Intel PIII 550) with Spot Advanced software, and further processed with Scion Image and Origin software packages. Capillary coating Capillaries were coated following a method described by Hjerten.24 In brief, uncoated fused-silica capillaries of 360 µm o.d. × 350 µm i.d. (Yongnian Optical Fiber Factory, Hebei Province, China) were flushed sequentially with 0.1 M NaOH, H2O and MeOH for 2 h each, and then dried with pure N2. After the pretreatment mentioned above, the inner surface of the capillaries was derivatized with MAPS (γmethacryloxypropyltrimethoxysilane), a bifunctional-group reagent, in a 1:1 MeOH solution and left at room temperature overnight. The capillary was then filled with a centrifuged and degassed 4% acrylamide solution containing 0.1% APS (Ammonium persulfate) and 0.1% TEMED (N,N,N′,N′tetramethylethylenediamine) overnight to form a linear polyacrylamide derived layer on the inner capillary wall. Next, the capillary was rinsed with H2O for 1 h to push any excessive polyacrylamide (not attached) out of the capillary. Finally, the capillary was filled with 37% formaldehyde (pH 10) for crosslinking the polyacrylamide. After 12 h, the capillary was flushed with H2O for 2 h and both ends were dipped into H2O for storage. Enzymatic reaction The recombinant plasmid was digested with HindIII (Promega) for 2 h at 37˚C. A restriction enzymatic reaction was performed in a total volume of 20 µL containing 1 µL HindIII (10 U/µL), 0.2 µL acetylated BSA (Bovine Serum Albumin) 2
Results and Discussion Optimization of CE method The aim of this work was to construct a capillary electrophoresis (CE) method with a non-gel sieving medium for identifying the orientation of DNA fragments in recombinant plasmids. To achieve such a purpose, the CE separation conditions should first be optimized. There are many factors that govern the electrophoretic separation quality. Because such a topic has been detailed by others,14–22,25,26 only a brief description is given here. In our work, a cellulose derivative solution of hydroethylcellulose (HEC) was selected as the sieving matrix because of its non-uniform chain length, which would be preferred for resolving DNA of wide range of lengths. The concentration of HEC has been varied from 0.4 to 2.0% to evaluate the sieving matrix. A concentration of 1.2% HEC (w/v) proved to be the best while separating a DNA marker of φX174/HaeIII restriction fragments. The time consumption dramatically increased with a higher HEC concentration, due to shrinkage of the sieving cores and an increase in the running buffer viscosity. Although the use of a capillary allows one to apply an electric field strength of up to 500 – 1000 V/cm in electrophoresis, a relatively lower electric field strength has been considered as a more practical choice for DNA separation. Our experimental data have shown that the best resolution with a moderate separation speed could be achieved under 200 V/cm, while the electric field strength ranged from 175 to 350 V/cm. A higher electric field strength provided a quicker separation speed, but the resolution became worse. In this work, an electrokinetic injection mode was employed due to its simplicity to the self-constructed instrument. Furthermore, a better resolution could be achieved due to a stacking effect if the ionic strength of the samples was lower than that of the sieving solution. When the injection voltage was fixed at –6 kV, an injection time of 5 s was chosen as the optimized parameter. For a lower sampling time, for example 2 s, the reproducibility of the sampling amount could not be guaranteed. Separation of DNA marker Based on the above-mentioned investigation, three DNA markers, including φX174/HaeIII, pBR322/HaeIII and λDNA/ HindIII, were tested under the optimization conditions (buffer, 1 × TBE containing 1 µM YO-PRO-1 and 1.2% HEC; applied electric field strength, 200 V/cm; electrokinetic sampling: time, 5 s; voltage, –6 kV). Figure 1 shows their electropherograms. It could be found that a good resolving power has been achieved for the DNA of short and medium length. Especially, a single base difference of 123 bp/124 bp has been discriminated in the separation of the DNA standard, pBR322/HaeIII. Relationship between fragment size and mobility The data obtained from Fig. 1 were used for evaluating the relationship between the logarithm of the fragment size and the electrophoretic mobility. Figure 2 shows the relationship that appears as a reversed sigmoid shape, which was in good agreement with the relationship in conventional SGE.27 It indicated that there might be very similar mechanism governing the electrophoretic separation in both electrophoretic methods. It has been widely accepted in SGE that the electrophoretic
ANALYTICAL SCIENCES NOVEMBER 2001, VOL. 17
1255
Fig. 2 Relationship between DNA Conditions are the same as Fig. 1.
size
and
the
mobility.
Fig. 1 Electropherograms of DNA markers. Buffer, 1 × TBE containing 1 µM YO-PRO-1 and 1.2% HEC, pH 8.3; capillary, 64 cm (total length) × 48 cm (effective length) × 75 mm i.d. × 360 µm o.d. with a polyacrylamide coated inner wall; electric field strength, 200 V/cm; injection, electrokinetic mode, 5 s; temperature, 25˚C.
mobility of DNA fragments of short length is based on the Ogston sieving mechanism, while the electrophoretic mobility of DNA fragments of medium length could be accounted for by a reptation mechanism. However, for DNA fragments of large length, the dependence of the mobility on size was very small. Their migration behavior could be deciphered by a bias reptation mechanism. The mechanisms mentioned above are currently being adopted in CE with non-gel sieving media. In Fig. 2, zone “a” of the curve was dominated by the Ogston mechanism, while zone “b” and zone “c” were governed by reptation or a bias reptation mechanism, respectively.
Fig. 3
Schematic description of enzyme approaches.
Identification of DNA orientation We recently cloned a new gene, A1, from a model plant Arabidopsis Thaliana (AT), which encoded a homology of calcium/calmodulin-dependent protein kinase 1. To further understand its function in cellular signal transduction, A1 was transferred into plant cells. As depicted in Fig. 3, a cyclic plasmid, pET32 a (+), was first cut into linear DNA at its poly-
1256
ANALYTICAL SCIENCES NOVEMBER 2001, VOL. 17 the lengths of the other two DNA fragments, it is a supercoil. The mechanism governing its migration behavior is quite different than that which the linear DNA depended on. The electrophoretic mobility of the supercoil could not be used for evaluating its size based on the relationship of the molecular size and the mobility established from the data of linear DNA. It was thus not surprising that it could migrate faster than a linear DNA fragment having a smaller size. Figure 4C shows another kind of result that proves that the sample belongs to the reversed orientation of the target gene. It would cause the expression of antisense of the target gene, which, in turn, would result in an inhibition of the translation of the protein encoded by A1.
References
Fig. 4 Electropherograms for identifying the orientation of gene A1 in recombinant plasmids. A, from right orientation; B, from uncompletely cleaved right orientation; C, from a reversed orientation.
cloning site by an endonuclease, BamH 1. Then, A1 was linked onto the plasmid end to end by the DNA ligase. There would be two ways for this ligation, as shown in Fig. 3 (a-a′, b-b′ or a-b′, b-a′). Consequently we would obtain two DNA products with difference in orientations. Also, the two products would have the same probability, theoretically. Because the ligation products were supercoils with the same molecular weight, it is impossible to identify them using the electrophoretic method. To discriminate the orientations of DNA fragments, CE or SGE should be conjugated with some other approaches, an enzymatic solution, for instance. If we could find two cleavage sites of an endonuclease, one in the plasmid pET32 a (+) and another in the target gene A1, there would be two DNA fragments formed due to cleavage of the endonuclease. Because the orientations of the target gene A1 were opposite in the two ligation products, the formed DNA fragments would be quite different in length, as indicated in Fig. 3. Thus, this difference in the DNA length could be feasible to be rapidly analyzed by CE. In this work, an endonuclease HindIII was selected to achieve such a purpose. Figure 4 gives electropherograms of three samples after treatments with the endonuclease HindIII. Figure 4A shows the result of a sample by CE, which proved to be the right orientation that would lead to an over-expression of the calcium/calmodulin-dependent protein kinase 1 encoded by A1. Regarding Fig. 4B, the sample was partially cleaved uncompletely by the endonuclease. The peak marked by an asterisk belongs to the un-cleaved cyclic DNA. Although the molecular weigh or length of this cyclic DNA equals the sum of
1. Y. P. Qi, “Genes and Principle of Gene Manipulation”, 1998, Wuhan University, Wuhan. 2. J. Sambrook, E. F. Fritsch, and T. Maniatis, “Molecular Cloning: A Laboratory Manual (2nd Ed.)”, 1989, Cold Spring Harbor Laboratory. 3. S. N. Krylov and N. J. Dovichi, Anal. Chem., 2000, 72, 111R. 4. A. M. Krstulovic, J. Chromatogr. B, 1997, 697, 1. 5. J. P. Landers, Electrophoresis, 1997, 18, 1709. 6. Z. El Rassi, Electrophoresis, 1997, 18, 2123. 7. C. Heller, J. Chromatogr. A, 1997, 806, 1. 8. B. R. McCord, Electrophoresis, 1998, 19, 1. 9. G. Bonn, J. Gerstner, and P. Oefner, J. Chromatogr. A, 1998, 806, 1. 10. E. S. Yeung and Q. Li, in “High Performance Capillary Electrophoresis—Theory, Techniques and Applications”, ed. M. G. Khaledi, 1998, Wiley-Inter-Science, New York, 767. 11. H. Drossman, J. A. Luckey, A. J. Kostichka, J. D. Cunha, and L. M. Smith, Anal. Chem., 1990, 62, 900. 12. D. A. McGregor and E. S. Yeung, J. Chromatogr., 1993, 652, 67. 13. M. Zhu, D. L. Hansen, S. Burd, and F. Gannon, J. Chromatogr., 1989, 480, 311. 14. D. N. Heiger, A. S. Cohen, and B. L. Karger, J. Chromatogr., 1990, 516, 33. 15. E. N. Fung, H. Pang, and E. S. Yeung, J. Chromatogr. A, 1998, 806, 157. 16. F. Han, B. H. Huynh, Y. Ma, and B. Lin, Anal. Chem., 1999, 71, 2385. 17. E. Carrilho, M. C. Ruiz-Martinez, J. Berka, I. Smirnor, W. Goetzinger, A. W. Miller, D. Brady, and B. L. Karger, Anal. Chem., 1996, 68, 3305. 18. H. Zhu, S. M. Clark, S. C. Benson, H. S. Rye, A. N. Glazer, and R. A. Mathies, Anal. Chem., 1994, 66, 1941. 19. A. E. Barron, W. M. Sunada, and H. W. Blanch, Electrophoresis, 1996, 17, 744. 20. N. Iki, Y. Kim, and E. S. Yeung, Anal. Chem., 1996, 68, 4321. 21. W. L. Tseng, M. M. Hsieh, S. J. Wang, and H. T. Chang, J. Chromatogr. A, 2000, 894, 219. 22. N. Chen, L. Wu, A. Palm, T. Srichaiyo, and S. Hjerten, Electrophoresis, 1996, 17, 1443. 23. Y. Baba, J. Chromatogr. B, 1996, 687, 271. 24. S. Hjerten, J. Chromatogr., 1985, 347, 191. 25. K. Ueno and E. S. Yeung, Anal. Chem., 1994, 66, 1424. 26. Y. Kim and M. D. Morris, Anal. Chem., 1994, 66, 1168. 27. O. J. Lumpkin, P. Dejardin, and B. H. Zimm, Biopolymers, 1985, 21, 1573.
