Anal. Chem. 1994,66, 4382-4383
Labeling of Double-Stranded DNA by ROX-Dideoxycytosine Triphosphate Using Terminal Deoxynucleotidyl Transferase and Separation by Capillary Electrophoresis Daniel Figeys, Annika Renborg, and Norman J. Dovichi* Department of Chemistry, University of Alberta, Edmonton, Alberta T6G 2G2, Canada
Terminal transferase is used to add a single fluorescently labeled dideoxynucleotide to double-strandedDNA prepared by restriction endonuclease action on a bacteriophage. The product is separated by capillary electrophoresis with both hydroxypropylmethylcellulose and non-crosslinked polyacrylamide. The reaction products generate single peaks for each fragment with hydroxypropylmethylcellulose. However, the higher resolution separation produced by non-cross-linkedpolyacrylamide shows that the product contains two componentsfor each restriction digest fragment. This labering technique should be useful in restriction fragment length polymorphism studies. Capillary electrophoresis is a powerful technique for the rapid separation of minute amounts of double-stranded DNA Sensitive detection has been based on laser-induced fluorescence. Intercalating dyes, such as Thiazole Orange, POPO-3, YOYO-3, and YOYO-1, have been used to detect doublestranded DNA separated by capillary While intercalating dyes are very convenient to use, they suffer from one important limitation. Because there is a rapid equilibrium between free and bound dye, it is not possible to label selectively different strands of DNA with different dyes. This limitation is particularly acute in capillary electrophoresis, where it is very useful to include a size marker along with the unknown DNA for analysis in a single capillary; of course, the throughput of each capillary is also increased if several samples, labeled with different fluorescence label, are analyzed simultaneously. We have attempted to use different intercalating dyes to label double-stranded DNA originating from two different restriction endonuclease digests; inevitably, exchange of the dyes was observed, resulting in the inability to distinguish the fluorescence from the two digests. As a further property of intercalating dyes, the structure and stiffness of the DNA are changed, modifying the mobility of the oligonucleotide. Instead, it appears necessary to label covalently DNA with fluorescent dye; the covalently bound dye is unable to exchange between strands, allowing analysis of two (or more) different analytes simultaneously in a single capillary. While it is possible to use fluorescently labeled primers in PCR generation of double-stranded DNA, these primers are expensive. Labeled primers are not useful in analysis of restriction endonuclease digests. (1) Schwartz, H. E.;Welder, K. J. Anal. Chem. 1992, 64, 1737-1740. (2) Figeys, D.; Aniaga, E.; Renborg, A; Dovichi, N.J. J. Chromatogr, 1994, 669, 205-216.
4382
Analytical Chemistry, Vol. 66, No. 23, December 1, 1994
We report the use of terminal deoxynucleotidyl transferase to label covalently both blunt and sticky-end DNA fragments with a fluorescently labeled dideoxynucleotide. Terminal deoxynucleotidyl transferase is widely used for 3'-end labeling of DNA or RNA with radioisotopically labeled nucleotide^.^*^ A fluorescently labeled deoxynucleotide has been used to label RNA with terminal deoxynucleotidyl tran~ferase.~ Trainor and Jensen used succinylfluorescein-labeled dideoxynucleoside triphosphate to label a 17-base-pair oligonucleotide.6 Igloi and Schiefermayr used terminal deoxynucleotidyl transferase to incorporate a fluorescently labeled dUTP to a single-stranded primer.7 However, to the best of our knowledge, terminal deoxynucleotidyl transferase has not been used to fluorescently label restriction fragment digests. EXPERIMENTAL SECTION
One microliter of DNA [@X174 RF/HueIII(Gibco BRL), 0.6 pg/pL, or M13mp18 RFl(Sigma)/TuqI(Gibco B E ) , 0.045 pg/ pL1, 4 pL of a 5x reaction buffer (Boehringer Mannheim), 1pL of ROX-ddCTP (90 pmol, ABI),2 pL of CoClz (2.5 mM, Boehringer Mannheim), 10p L of water, and 1pL of terminal deoxynucleotidyl transferase (10 units, Boehringer Mannheim) from calf thymus are mixed together. The mixture is incubated at 37 "C for 1 h. EDTA is added to terminate the reaction. The sample is ethanol precipitated and resuspended in 10 pL of 1x TBE (0.54 g of Tris, 0.275 g of boric acid, and 0.100 mmol of disodium EDTA, diluted to 50 mL with deionized water). The labeled product was separated by capillary electrophoresis in a locally constructed i n ~ t r u m e n t .A~ ~yellow ~ ~ ~ He-Ne laser (594 nm) is used for excitation of fluorescence. The separation was performed with 0.4%hydroxypropylmethylcellulose at an electric field of 100V/cm. The 32-pm4.d. capillary is 28.6 cm long. RESULTS AND DISCUSSION
We used ROX-dideoxycytosine triphosphate (Applied Biosystems Division of Perkin-Elmer) to labeled blunt and sticky-end restriction fragments. We successfully labeled ax174 RF/HueIII (3) Tu,C.-P. D.; Cohen, S. N.Gene 1980, 10, 177-183. (4) Roychoudhury, B.; Jay, E.;Wu, R Nucleic Acids Res. 1976, 3, 863-877. (5) Dirks, R W.; Van Gijlswiljk, R. P. M.; Vooijs, M. A,; Smit, A. B.; Bogerd, J.; Van Minnen, J.; Raap, A. IC;Van Der Ploeg, M. Exp. Cell Res. 1991, 194, 310-315. (6) Trainor, G. L.; Jensen, M. A. Nucleic Acids Res. 1988, 16, 11846. (7) Igloi, G. L.; Schiefennayr, E.BioTechniques 1993, 15, 486. (8) Chen, D. Y.; Harke, H. R; Dovichi, N. J. Nucleic Acids Res. 1992,20,48734880. (9) Chen, D. Y.; Swerdlow, H. P.; Harke, H. R.; Zhang, J. Z.; Dovichi. N.J . J, Chromatogr. 1991, 559, 237-246.
0003-2700/94/0366-4382$04.50/0 0 1994 American Chemical Society
234 118
1
310/ 1078
872
6
8
10
12 14 16 Elution time (Minutes)
18
20
Figure 1. Separation of DX174 RFIHaelll on hydroxypropylmethylcellulose, 0.4%, at an electric field of 100V/cm. The running buffer is 1 x TBE plus 0.1 M NaCI.
and Ml3mpl8ITuqI restriction fragments. We have reported the use of hydroxypropylmethyl cellulose for the separation of double stranded DNA (2). This material is very easy to use and produces good separations. Figure 1 shows the separation of ax174 RF/ Hue111 fragments on hydroxypropylmethylcellulose. Similar results were obtained with MlJmplB/TuqI. While hydroxypropylmethylcellulose is convenient for analysis of double-stranded DNA, non-cross-linked polyacrylamide solutions provide higher resolution separations. Figure 2 presents the separation of ax174 RF/HueIII fragments on polyacrylamide. Each fragment generates a major peak and a smaller satellite peak; presumably, the first peak is due to fragments that have incorporated a single label at one 3’-end, and the second peak is associated with fragments that have been labeled at both ?-ends. For the largest peaks, the resolution presumably is insufficient to resolve the satellite peaks. Clearly, more work is required to minimize the formation of the satellite peaks. As suggested by a reviewer, the decrease in the relative amount of dideoxynucleotide should reduce formation of the satellite peaks. The use of fluorescent dideoxynucleotides is attractive for labeling of double-stranded DNA. In particular, the labeling
6
8
10
12
14
16
18
20
Elution time (Minutes)
Figure 2. Separation of mX174 RFIHaelll on polyacrylamide(5%T, O%C) at an electric field of 200 V/cm. The running buffer is 1x TBE plus 0.1 M NaCI.
procedure will be of value in the analysis of restriction endonuclease digested DNA; alternative technology based on the use of intercalating dyes does not easily allow analysis of mixed samples. There are at least four different dideoxynucleotides labeled with unique fluorescent probes. In principle, a mixture containing three samples plus an internal standard can be analyzed simultaneously in a single capillary. ACKNOWLEDGMENT This work was supported in part by the US. Department of Energy Human Genome Initiative Grant No. DEFG02-91ER61123. Support by DOE does not constitute an endorsement of the views expressed in this article. This work was also supported by the Natural Sciences and Engineering Research Council of Canada (NSERC). D.F. acknowledges a graduate fellowship, and N.J.D. acknowledges a Steacie fellowship from NSERC. Received for review May 26, 1994. Accepted September 12,1994.@ @
Abstract published in Advance ACS Abstracts, October 15, 1994.
Analytical Chemistry, Vol. 66,No. 23, December 1, 1994
4383
