Bioconjugate Chem. 1998, 9, 671−675
671
Molecular Hybridization Probes Prepared with 4-Aminooxybutylamine Vyacheslav A. Adarichev, Sergey M. Kalachikov, Anna V. Kiseliova, and Grigory M. Dymshits* Institute of Cytology and Genetics SD RAS, 10 Lavrentyeva Avenue, Novosibirsk 630090, Russia. Received January 14, 1998; Revised Manuscript Received September 15, 1998
A versatile method is described for preparing nonradioactive DNA probes for molecular hybridization. The method is based on the transamination reaction of double-stranded DNA with 4-aminooxybutylamine (ABA). To optimize the procedure for obtaining stable and sensitive hybridization probes, time of modification, pH, and reaction temperature were varied. The optimal reaction conditions allowed the preparation of nonradioactive-labeled DNA probes that met demands for optimal length, modification degree, stability, and sensitivity. The use of 4-aminooxybutylamine as a bifunctional reagent for DNA modification allowed the possibility of choosing an appropriate reporter group: biotin or one of the fluorochromes. For probes carrying biotin, a high specificity and high sensitivity detection limit of 1 pg of target DNA were demonstrated. In addition, the applicability of probes carrying fluorochromes for multicolor direct fluorescence in situ hybridization was described.
INTRODUCTION
Nucleophilic reagents are broadly used to introduce nonradioactive labels into polynucleotides: bisulfite (Lebacq et al., 1988), aliphatic amines (Turchinskii et al., 1988), hydrazides (Reisfeld et al., 1987; Turchinskii et al., 1989; Karpyshev et al., 1990), and substituted hydroxylamines (Adarichev et al., 1987; Brosalina et al., 1987). Within this range of chemicals, hydroxylamine and its derivatives display the highest rate of nucleophilic substitution and best stability (Kochetkov et al., 1971, 1972). Therefore, we tested 4-aminooxybutylamine (ABA)1 as a carrier of nonradioactive labels. This bifunctional agent carries an aminooxy group on one end and an amino group on the other. On the first step of DNA modification (Scheme 1), the aminooxy group reacts selectively with cytosine, and the amino group allows a variety of reporter compounds such as biotin and different fluorochromes to be attached on the second step. We reported earlier that ABA can be used for introduction of amino groups into extended double-stranded polynucleotides (Adarichev et al., 1987). The aim of the present work was to determine the optimal conditions for the reaction of DNA with ABA in order to obtain highly sensitive nonradioactive probes. The probes were characterized for their application in molecular hybridization.
Scheme 1. Synthesis of Biotin and/or FluoresceinLabeled DNA Probes in Two-Step Modification Procedurea
EXPERIMENTAL PROCEDURES
ABA was synthesized by Vector Inc., Koltsovo, Russia. Streptavidin-alkaline phosphatase conjugate was purchased from Boehringer Mannheim, Switzerland; agarose, n-nitrotetrasolium blue (NBT), 5-bromo-4-chloro-3indolyl phosphate toluidine salt (BCIP), N-hydroxysuccinic ether D-biotinylaminocaproate, DMSO, and salmon sperm DNA were from Sigma and SDS, RNAse, tetramethyl* To whom correspondence should be addressed. Tel: (3812) 35-4743. Fax: (3812) 35-6558. E-mail:
[email protected]. 1 Abbreviations: 1×SSC, 0.15 M NaCl, 0.015 M sodium citrate, pH 7.0; ABA, 4-aminooxybutylamine; SDS, Na dodecyl sulfate; kb, kilobases.
a At Step 1, cytosine of DNA reacts with ABA, and the major product of ABA incorporation into DNA is shown. The structure based on temperature, concentration, and pH dependencies of cytosine modification with O-alkyl derivatives of hydroxylamine (Budowsky et al., 1971; Kochetkov et al., 1971, 1972; Petrenko and Spirin, 1982). At Step 2, ABA-DNA reacts with activated derivatives of biotin or fluorochromes (R).
rhodamine isothiocyanate, fluorescein isothiocyanate isomer I were from Fluka, Switzerland.
10.1021/bc980007u CCC: $15.00 © 1998 American Chemical Society Published on Web 10/31/1998
672 Bioconjugate Chem., Vol. 9, No. 6, 1998
Modification of Polynucleotides. Step 1. Reaction of DNA with ABA. Chemical modification of polynucleotides was performed with 20 µL of 1 µg/µL plasmid DNA and 10 µL of 1.0 M ABA at pH 3.6. The final volume was adjusted with water to 40 µL; the mixture was incubated for 1-40 min at 65, 80, or 100 °C. Acidity of ABA solution was adjusted to the required pH value by titration at 65 or 80 °C. Reaction pH at 100 °C was calculated based on pH at 65 °C and dependency of the ABA ionization constant with temperature. The value ∆Ka/∆C was -0.03 deg-1 for 1.0 M ABA solution and was not affected by temperature over the range 30-80 °C. Removal of an uncoupled reagent was performed on Sephadex G-25 gel filtration by elution with 20 mM NaHCO3. DNA solution was concentrated by successive extractions with n-butanol and diethyl ether. Ether was removed by vacuum. Modified DNA was stored at -20 °C. Step 2. Attachment of reporter compounds to modified polynucleotides was done as described earlier (Frumgarts et al., 1986). The reaction mixture consisted of 10 mM of N-hydroxysuccinic ether D-biotinylaminocaproate, tetramethylrhodamine isothiocyanate, or fluorescein isothiocyanate and 0.1 M NaHCO3, 0.2 µg/µL ABA-DNA, and 10% DMSO. After incubation for 2 h at 37 °C, the mixture was loaded onto Sephadex G-25 gel and eluted with 20 mM NaHCO3 to remove uncoupled reagents. DNA samples labeled with fluorescein were characterized by absorption at 260 and 490 nm to calculate the ratio of bases carrying fluorochrome. Labeled DNA was stored at -20 °C. To evaluate the percentage of nucleotide bases of DNA carrying ABA or biotin (degree of modification), an aliquot of ABA-DNA after first step of modification was labeled with fluorescein. The degree of modification for fluorescein-labeled DNA was determined as a percentage of labeled bases over all bases using the 490/260 nm absorption ratio. The extinction coefficient of fluorescein in 20 mM NaHCO3 was 42 000 M-1 cm-1 at 490 nm and 14 400 M-1 cm-1 at 260 nm, and the extinction coefficient of DNA was 6500 M-1 cm-1 at 260 nm. The sensitivity of biotin-labeled DNA probes was estimated by homologous dot-blot hybridization. Decreasing amounts of target DNA were immobilized onto nylon membrane, and DNA probe was labeled in a twostep procedure prior to hybridization with DNA on membrane. The probe sensitivity (in picograms) was defined as a lowest detectable amount of target DNA. Distribution of labeled DNA fragments was confirmed by agarose gel electrophoresis in 30 mM NaOH. Fluorescein-labeled DNA in agarose gel was photographed under ultraviolet illumination using “Photodyne” camera with a green filter. Biotin-labeled DNA was transferred from agarose gels under vacuum onto nylon membranes and immobilized by ultraviolet illumination (Kalachikov et al., 1992). Biotin was detected with streptavidinalkaline phosphatase conjugate as described earlier (Langer et al., 1981). Chromosome In Situ Hybridization Using Fluorochrome-Labeled Probes. Slides with chromosomes of mink, mouse, and vole Microtus arvalis were kindly provided by Dr. S. M. Zakian and Dr. O. L. Serov (Institute of Cytology and Genetics, Novosibirsk, Russia). Slides were denatured with 70% formamide in 2×SSC at 70 °C for 2 min and immediately transferred into 70, 80, and 96% ethanol series for 5 min each at 2 °C, and slides were then air-dried. Fluorochome-labeled probe (1 µg) was combined with 2 µg of sheared salmon sperm DNA and 20 µg of sheared Escherichia coli DNA in 40
Adarichev et al.
µL of hybridization buffer containing 50% formamide and 10% dextran sulfate in 2×SSC. The mixture was denatured at 90 °C for 3 min, immediately chilled on ice for 5 min, and then added to the slides. Hybridization was carried out overnight in a moist chamber at 37 °C. Slides were washed twice in 50% formamide in 2×SSC, twice in 2×SSC at 42 °C, stained with propidium iodide, and mounted in 5 mM p-phenylendiamine, 70% glycerol, and 2×SSC. Fluorescent signal was detected with luminescent microscope “Lumam I-3” (LOMO, Russia). RESULTS AND DISCUSSION
Specifying the Reaction Conditions. Hydroxylamine and its derivatives possess a high specificity for single-stranded nucleic acids. Phage M13 single-stranded DNA can be modified by 10% for 20 h in 1.0 M ABA, pH 4.5, at 37 °C. No breaks in DNA were detected in the course of modification, and the corresponding biotin probes displayed a high sensitivity upon homologous dotblot hybridization. Under identical conditions, doublestranded DNA can be modified by less than 1% (data not shown). To denature DNA and promote transamination, the temperature was raised. The rate of ABA incorporation was sharply increased with the reaction temperature, and DNA probes obtained at 100 °C demonstrated a higher degree of modification with ABA/biotin compared with probes obtained at 80 and 65 °C (Figure 1). Nevertheless, despite the significantly higher quantity of biotin moieties, the probes produced at 100 °C demonstrated the least sensitivity among all tested samples. Thus, 80 °C was the optimal temperature for the first step of DNA modification due to the combination of high rate of ABA incorporation and high sensitivity of probes in homologous dot-blot hybridization. DNA modification with ABA at 80 °C (Figure 2) showed a steady increase in ABA incorporation with time. However, the results from homologous dot-blot hybridization of corresponding biotinylated DNA probes, produced under these conditions, demonstrated that sensitivity was maximal for probes incubated for 5-10 min with ABA, and longer incubation decreased the hybridization signal. In addition, the portion of the probe bound nonspecifically to the membrane increased under extended incubation. The level of DNA probe sensitivity achieved after 5-10 min of modification was greater than 1 pg. This level of sensitivity was comparable to that of DNA probes prepared by other nonradioactive-labeling methods, both chemical and enzymatic (Langer et al., 1981; Reisfeld et al., 1987; Lebacq et al., 1988; Turchinskii et al., 1989; Karpyshev et al., 1990). Characterization of the Probes Produced by DNA Transamination. It was shown that O-alkylhydroxylamines demonstrate high specificity to cytosine, does not react with guanine and thymidine, and reacts minimally with adenine nucleus (Budowsky et al., 1971a,b; Budowsky et al., 1972; Kochetkov et al., 1971, 1972). Despite ABA belonging to O-alkylhydroxylamines, its reactive specificity may differ under particular conditions we used for modification. To determine the reactivity of ABA to different DNA bases, we incubate the reagent with single-stranded homopolynucleotides under conditions optimal for DNA modification (0.25 M ABA pH 3.6, 80 °C, 10 min). Indeed, ABA showed reactivity for each of the DNA bases. The portion of modified cytosine among all the four bases was about 70%, but ABA reacted also with guanine (16%), adenine (10%), and thymidine (4%). The resulting ratio of rates (G > A > T) was identical for DNA depurination upon heating in acidic
Modification of DNA with 4-Aminooxybutylamine
Bioconjugate Chem., Vol. 9, No. 6, 1998 673
Figure 1. Time course of the ABA incorporation into double-stranded DNA at different temperatures: (A) 100 °C; (B) 80 °C; (C) 65 °C, on left. Homologous dot-blot hybridization of the probes modified for 10 min, on right.
Figure 2. Time course of the ABA incorporation into plasmid DNA at 80 °C, on left. Homologous dot-blot hybridization of the produced probes, on right. Labels 2, 6, 15, 30 min, duration of the modification.
conditions (Kochetkov et al., 1971, 1972). Obviously, detachment of DNA bases followed by coupling the ABA with C1-deoxyribose aldehydes groups might explain homopolynucleotides reactivity. Depurination at the C1 atom can lead to disintegration of polynucleotides and subsequent changes in the hybridization properties of the probe. For that reason, DNA probes obtained in the two-step labeling procedure were tested by gel electrophoresis under alkali conditions, which allowed us to detect all anucleated sites. Plasmid DNA was incubated with ABA and modified with biotin. DNA sample was fractionated on alkaline agarose gel and transferred onto nylon membrane. Following these procedures, the 6.0 kb plasmid DNA demonstrated some degree of degradation (Figure 3). Nevertheless, a final length of 1-5 kb remained optimal for hybridization. A similar size distribution pattern was also obtained for the fluorescein-labeled DNA. The optimal degree for DNA probe labeling was studied by homologous dot-blot hybridization using plasmid DNAs carrying 0.2-8% biotin. Target DNA was spotted onto nylon membrane in decreasing amounts, and each biotinylated DNA sample was hybridized with target under standard conditions. Dependence of sensitivity at
Figure 3. Size distribution for DNA probes labeled with ABA and biotin after gel electrophoresis in alkaline conditions.
674 Bioconjugate Chem., Vol. 9, No. 6, 1998
Adarichev et al.
Figure 4. Sensitivity of biotinylated DNA probes with different degree of modification in homologous dot-blot hybridization. Figure 6. Direct multicolor fluorescent in situ hybridization with a mixture of mink and mouse cell nuclei. The probe for mink carried tetramethylrhodamine (orange), the probe for mice carried fluorescein (green).
Figure 5. ABA-modified DNA probes in fluorescentin situ hybridization with chromosomes of the vole Microtus arvalis. Transaminated and labeled with fluorescein plasmid carrying species-specific repeated sequences was used as a probe. Hybridization signal was detected by fluorescent microscopy (greenyellow). Chromosomes were contrasted with propidium iodide staining (red).
degree of modification demonstrated plateau started from 1.5% biotinylated bases (Figure 4). Biotinylated DNA probes that carried 1.5-8% biotin demonstrated equal sensitivity, which dropped abruptly if DNA probes carried less than 1% biotin. Thus, the maximum probe sensitivity for biotin-ABA label was at 1.5% of modified DNA bases. Higher degree of modification resulted in higher density of biotin labels on the DNA strand, and obviously, steric problems arose when the sizable streptavidinalkaline phosphatase conjugate interacted with the biotin residues hindering increased sensitivity. In the case of fluorochrome-labeled probes, the dependence was also asymptotic, but maximal sensitivity was reached at 4% modified DNA bases. In the latter case, the mechanism of quenching of fluorescent signal was responsible for sensitivity hindering (Frumgarts et al., 1986). The application of ABA-modified DNA probes was demonstrated for in situ hybridization. Plasmid DNA carrying species-specific sequences were modified with ABA, and either biotin or fluorescein was attached. The
species-specific repeat was localized in the heterochromatin arm of the sex chromosomes of the vole Microtus arvalis by direct signal detection from fluorescein-labeled DNA probes (Figure 5). Nonradioactive hybridization of the biotinylated plasmid demonstrated identical chromosomal localization (data not shown). The described two-step labeling procedure with ABA as the first modifying agent allows the choice of the most suitable label for the task. In the frame of one method, different fluorochromes can be attached to DNA probes opening the possibility of simultaneous hybridization of several probes carrying fluorochromes with different emission spectra. Figure 6 presents the results of multicolor fluorescent hybridization with a mixture of mink and mouse cells. Total mink DNA, as a first DNA probe, was labeled with ABA and tetramethylrhodamine (orangeyellow nuclei), and total mouse DNA, as a second probe, carried ABA-fluorescein label (green nuclei). The described procedure for the preparation of molecular hybridization probes does not require enzymes or unstable compounds. In one round of two-step modification, a large quantity up to hundred microgram of fluorochrome- or biotin-labeled probe can be produced starting from double-stranded plasmid DNA. These probes can be used in all fields of molecular hybridization. ACKNOWLEDGMENT
The authors acknowledge the helpful comments of Drs. E. I. Budowsky and B. P. Ulanov as well as collaboration of Drs. A. S. Grafodatsky, S. M. Zakian, G. A. Zayniyev, L. P. Zakharenko, L. P. Matiakhina, O. V. Sablina, and O. L. Serov and the support from Programs “Human Genome” and “Frontiers in Genetics” of Russian Academy of Science. The editorial assistance of Vladimir Filonenko is greatly appreciated. LITERATURE CITED Adarichev, V. A., Dymshits, G. M., Kalachikov, S. M., Pozdnyakov, P. I., and Salganik, R. I. (1987) DNA Carrying Aliphatic Amino Groups and Use of Its Fluorescent Derivative as a Probe in Molecular Hybridization. Bioorg. Khim. 13 (8), 1066-9.
Modification of DNA with 4-Aminooxybutylamine Brosalina, E. B., Vlasov, V. V., Grachev, S. A., and Demchenko, E. N. (1987) Construction of DNA-Probes and Their Immunochemical Detection Using Nonradioactive Hybridization Tests. Bioorg. Khim. 13 (12), 1644-54. Budowsky, E. I., Sverdlov, E. D., and Monastyrskaya, G. S. (1971a) Mechanism Of the Mutagenic Action Of Hydroxylamine. IV. Reaction Of Hydroxylamine And O-Methylhydroxylamine With Adenine Nucleus. Biochim. Biophys. Acta 246 (2), 320-8. Budowsky, E. I., Sverdlov, E. D., Shibaeva, R. P., Monastyrskaya, G. S., and Kochetkov N. K. (1971b) Mechanism of the Mutagenic Action of Hydrohylamine. III. Reaction of Hydroxylamine and O-methylhydroxylamine with the Cytosine Nucleus. Biochim. Biophys. Acta 246 (2), 300-19. Budowsky, E. I., Turchinsky, M. F., Domkin, V. D., Pogorelov, A. G., Pisarenko, V. N., Kusova, K. S., and Kochetkov, N. K. (1972) Mechanism of the Mutagenic Action of Hydrohylamine. VI. Reaction of Hydroxylamine with the Uracil Nucleus. Biochim. Biophys. Acta 277 (2), 412-27. Frumgarts, L. A., Kipriianov, S. M., Kalachikov, S. M., Dudareva, N. A., and Dymshits, G. M. (1986) Preparation of Fluorescent-Labeled DNA and Its Use as a Probe in Molecular Hybridization. Bioorg. Khim. 12 (11), 1508-13. Kalachikov, S. M., Adarichev, V. A., and Dymshits, G. M. (1992) Ultraviolet Cross-linking of DNA to Microporous Membranes. Bioorg. Khim. 18 (1), 52-62. Karpyshev, N. N., Bondarenko, Tlu., and Kipriianov, S. M. (1990) Evaluation of the Effectiveness of Hybridization Probes Prepared by Action of Hydrazides of Biotin Derivatives on DNA. Bioorg. Khim. 16 (5), 605-9. Kochetkov, N. K., Budowsky, E. I., Sverdlov, E. D., Simukova, N. A., Turchinskii, M. F., and Shibaev, V. N. (1971, 1972)
Bioconjugate Chem., Vol. 9, No. 6, 1998 675 Organic Chemistry of Nucleic Acids (N. K. Kochetkov, and E. I. Budowsky, Eds.) Plenum Press, New York, London. Langer, P. R., Waldrop, A. A., and Ward, D. C. (1981) Enzymatic Synthesis of Biotin-Labeled Polynucleotides: Novel Nucleic Acids Affinity Probes. Proc. Natl. Acad. Sci. U.S.A. 78 (11), 6633-7. Lebacq, P., Squalli, D., Duchenne, M., Pouletty, P., and Joannes, M. (1988) A New Sensitive Non-Isotopic Method Using Sulfonated Probes to Detect Picogram Quantities of Specific DNA Sequences on Blot Hybridization. J. Biochem. Biophys. Methods 15, 255-66. Petrenko, V. A., and Spirin, G. V. (1982) Reaction of cytidylic acid with O-delta-aminooxybutylhydroxylamine. Mol. Biol. (Moscow) 16 (3), 644-8. Reisfeld, A., Rothenberg J. M., Bayer, E. A., and Wilchek, M. (1987) Nonradioactive Hybridization Probes Prepared by the Reaction of Biotin Hydrazide with DNA. Biochem. Biophys. Res. Commun. 142 (2), 519-26. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual, 2nd ed. Cold Spring Harbor Laboratory Press, Plainview, NY. Turchinskii, M. F., Aibinder, E. I., and Sverdlov, E. D. (1988) Multiple Hybridization in Genome Analysis. 1. Reaction of Diamine and Bisulfite with Cytosine for the Introduction of Nonradioactive Markers into DNA. Mol. Biol. (Moscow) 22 (6), 1545-52. Turchinskii, M. F., Aibinder, E. I., Knorre, V. D., and Scherbo, S. N. (1989) Modification of DNA with Dihydrazide of Succinic Acid for Preparation of Hybridization Probes. Bioorg. Khim. 15 (10), 1341-5.
BC980007U
