- Email: [email protected]
Overcoming false negatives due to the genomic context in polymerase chain reaction amplification
J. Biochem. Biophys. Methods 40 (1999) 45–48
Overcoming false negatives due to the genomic context in polymerase chain reaction amplification E. Baldrich, N. Xamena, O. Cabre´ * ` ` ` Grup de Mutagenesi , Unitat de Genetica , Departament de Genetica i de Microbiologia, Edifici Cn, ` ` ), 08193 Barcelona, Spain Universitat Autonoma de Barcelona, 08193 Bellaterra ( Cerdanyola del Valles Accepted 3 April 1999
Abstract In some instances defined genomic regions are so poorly amplified that they seem to be unamplifiable. A protocol was developed which allows good PCR amplifications by the use of restriction digestion combined with the elution of a pool of restriction fragments of defined size range from agarose gel after electrophoresis. We describe the application of the method in the PCR amplification of a region of the white locus of Drosophila melanogaster that otherwise may be considered as a negative result. 1999 Elsevier Science B.V. All rights reserved. Keywords: Drosophila melanogaster; Mutation; PCR specificity; white locus
The polymerase chain reaction (PCR) is a very useful technique with multiple applications, but often generates multiple undesired products. These may be due to nonspecific annealing of primers at other targets or at similar sequences. To improve the results, changes in the PCR parameters are usually programmed and / or other PCR improved methods performed. Taken into account that the template sequence must be known to choose the adequate primers, some authors [1,2] have found that predigesting the DNA with restriction enzymes that cut outside the region to be amplified, a good amplification of the expected region can be obtained. We have observed a lack of amplification of defined regions depending on the analyzed strains, but with no apparent cause. These amplifications can be taken as *Corresponding author. Tel.: 1 34-93-5811-662; fax: 1 34-93-5812-387. ´ E-mail address: [email protected] (O. Cabre) 0165-022X / 99 / $ – see front matter 1999 Elsevier Science B.V. All rights reserved. PII: S0165-022X( 99 )00017-2
46
E. Baldrich et al. / J. Biochem. Biophys. Methods 40 (1999) 45 – 48
negative results but are, in fact, very poor amplifications behaving as false negatives. We present here a protocol to determine whether a lack of amplification is only a very poor amplification, not detectable by electrophoresis, that might be due to the genomic context. In our particular case we attempted to amplify a 646 base-pair (bp) segment of the white locus of Drosophila melanogaster in some light eye colour mutants appeared in our laboratory and some of its germinal and somatic revertants. This region includes the insertion site of copia retrotransposon characterizing white–apricot allele [3], also considered the zero point of the Levis coordinates [4]. The results obtained by Southern blot showed that the white locus seemed to be intact in all the analyzed mutants, but the region including the empty copia insertion site was not amplified by PCR. The primers were designed from the white locus sequence reported [5,6]. The direct primer (59GGGAATTCATACTTCGTTTTGT-39) anneals at position 1 396 and the reverse primer (59-GTTAGGGAGCCGATAAAGAG-39) anneals at position 2 250, of the Levis coordinates. The standard reactions were carried out in a final volume of 25 ml of reaction mixture containing 50 mM KCl, 10 mM Tris–HCl (pH 8.0), 2 mM MgCl 2 , 0.2 U Taq-polymerase (GIBCO-BRL), 0.15 mM each of dNTPs, 0.25 mM each of primers and 1 ml of DNA template (100–400 ng / ml). Standard PCR parameters were: 948C, 3 min; 29 cycles consisting of: 948C, 45 s; 608C, 45 s; 728C, 90 s; 728C, 5 min. Fig. 1a shows the result of different amplifications carried out are presented here as an example. Fig. 1b shows the Southern blot of the same electrophoresis hybridized with p2k plasmid as probe, to detect poor amplifications and discard nonspecific ones. The p2k plasmid is a 2 kb long BamHI–SacI restriction fragment of the white locus, that contains the sequence to be amplified, cloned in the pTZ-18 plasmid. The expected amplification of 646 bp segment is shown in Fig. 1a, lanes 2 and 4, corresponding to p2k plasmid and wild type strain Canton-S, respectively. Lanes 3 and 8 show controls of no amplification: pTZ-18, which does not contain the sequence to be amplified, (lane 3) and white–apricot mutant, where copia insertion prevents amplification (lane 8). When the amplification was performed with DNA from our mutants no amplification product could be observed, as shown in Fig. 1a, lane 9 for M115 mutant as a example. In order to determine whether the segment was not amplified or only poorly achieved, the PCR products were Southern blotted and hybridized with the p2k probe (Fig. 1b). The expected fragments could then be detected in all the studied mutants as can be seen in lane 9 for M115 mutant. This means that the 646 bp segment was being amplified but with a very low efficiency. Changes in the different amplification reaction conditions or in the mixture components did not improve the result. Amplifications using other primers annealing near the previously described ones (for instance the direct primer 59-ACGAATTGATTTTTAGGGAC-39 that anneals at position 1 745), gave the same negative result. On the other hand, reamplifications and seminested reactions (for instance using the internal direct primer 59-ATGAATTCTGGCAG-39 annealing at 1 385) resulted in a high number of faint bands of unspecific products (Fig. 1a; lane 5: Canton-S as control and lane 10: M115 mutant). In similar cases other authors have proposed a predigestion with restriction enzymes which cut outside the region to be amplified in order to detect viral DNA in human cells
E. Baldrich et al. / J. Biochem. Biophys. Methods 40 (1999) 45 – 48
47
Fig. 1. (a) Example of PCR amplification products obtained after different treatments of templates. Lanes 1 and 14: HaeIII digested fX174 RF DNA length marker. Lane 13: water control. Lanes 2 to 12: 4 ml of final amplification products obtained from the amounts of templates indicated as follows. Lane 2: p2k plasmid (1 ml, | 100 ng / ml). Lane 3: pTZ-18 plasmid (1 ml, | 100 ng / ml). Lane 4: Canton-S (1 ml, | 400 ng / ml). Lane 5: 1 ml of previously amplified DNA from Canton-S. Lane 6: | 400 ng of DNA from Canton-S, digested with BamHI and SacI. Lane 7: 1 ml restriction fragments of | 2 kb in size obtained from Canton-S DNA treated as described in the text. Lane 8: w a (1 ml, | 400 ng / ml). Lane 9: M115 mutant (1 ml, | 400 ng / ml). Lane 10, 1 ml of previously amplified DNA from M115. Lane 11: | 400 ng of DNA from M115, digested with BamHI and SacI. Lane 12: 1 ml restriction fragments of | 2 kb in size obtained from M115 mutant DNA treated as described in the text. (b) The electrophoresis of (a) after Southern blot and hybridization with p2k probe. Detection was carried out by a nonradioactive method (Boehringer).
[1] or human DNA in somatic cell hybrids [2]. We repeatedly found a lack of amplification when the DNA of our mutants and revertants were predigested with restriction enzymes (BamHI and SacI) in order to eliminate unspecific amplifications (Fig. 1a; lane 6: Canton-S as control and lane 11: M115 mutant). Since DNA amplifications of wild type Canton-S strain behaved as expected when used as control, we supposed that some kind of interaction with the genomic context could raise difficulties to the template amplification in our mutants and revertants. For this reason we decided to eliminate, as much as possible, the interferences between our target DNA and the rest of the genome. Therefore we developed a two step protocol. The first step is a digestion of genomic DNA with a restriction enzyme(s) that cuts outside the region to be amplified. The second step is a electrophoresis of digested DNA, followed by recovering the restriction fragments of approximately the desired size by elution from the agarose gel. To achieve this, 5 mg of the total genomic DNA from each sample was digested with BamHI and SacI, generating a 2 kb restriction fragment of the white locus that includes the 646 bp segment to be amplified. The digested DNAs and a DNA length marker were loaded on a 1.5% agarose gel, and run 5 h at 70–80 V. Taking as reference the length of the marker DNA bands, portions of gel containing restriction fragments of approximately 2 kb were cut off from each lane of the gel. The fragments were then eluted in a final
48
E. Baldrich et al. / J. Biochem. Biophys. Methods 40 (1999) 45 – 48
volume of 15–30 ml, using the commercial kit ‘Geneclean’ (Bio101). A 1-ml volume of each sample was used as substrate for PCR. The expected 646 bp fragment was successfully amplified in all the cases when the eluates were used as substrate to carry out the PCR reaction. The amplification was always efficient enough to be directly detected after running 2–4 ml on an agarose gel or, for instance, to be directly sequenced (Fig. 1a, lane 12). We think that the protocol proposed by us in this paper is useful to overcome false negatives in PCR amplification, and improves the methods previously described [1,2]. The previous electrophoretic isolation of target DNA of defined size cut by only one or two enzymes, allows to amplify the target DNA when annoying sequences are unknown and avoids the use of cocktail enzymes.
Acknowledgements ´ y Cultura’ (grant PB96This work was supported by the ‘Ministerio de Educacion 1138, DGES) and E.B. was supported by a doctoral fellowship (FPI) from the ‘Generalitat de Catalunya’.
References [1] Sharma JK, Gopalkrishna V, Das BC. A simple method for elimination of unspecific amplifications in polymerase chain reaction. Nucleic Acids Res 1992;20:6117–8. [2] Hatchwell E. Improving PCR specificity by multiple pre-digestion. Trends Genet 1996;12:460. [3] Bingham PM, Judd BH. A copy of the copia transposable elements very tightly linked to the w a allele at the white locus of D. melanogaster. Cell 1981;5:705–11. [4] Levis R, Bingham P, Rubin G. Physical map of the white locus of Drosophila melanogaster. Proc Natl Acad Sci USA 1982;79:564–8. [5] O’Hare K, Murphy C, Levis R, Rubin GM. DNA sequence of the white locus of Drosophila melanogaster. J Mol Biol 1984;80:437–55. [6] O’Hare K. EMBL / GenBank accession number X02974, 1986.
Overcoming false negatives due to the genomic context in polymerase chain reaction amplification E. Baldrich, N. Xamena, O. Cabre´ * ` ` ` Grup de Mutagenesi , Unitat de Genetica , Departament de Genetica i de Microbiologia, Edifici Cn, ` ` ), 08193 Barcelona, Spain Universitat Autonoma de Barcelona, 08193 Bellaterra ( Cerdanyola del Valles Accepted 3 April 1999
Abstract In some instances defined genomic regions are so poorly amplified that they seem to be unamplifiable. A protocol was developed which allows good PCR amplifications by the use of restriction digestion combined with the elution of a pool of restriction fragments of defined size range from agarose gel after electrophoresis. We describe the application of the method in the PCR amplification of a region of the white locus of Drosophila melanogaster that otherwise may be considered as a negative result. 1999 Elsevier Science B.V. All rights reserved. Keywords: Drosophila melanogaster; Mutation; PCR specificity; white locus
The polymerase chain reaction (PCR) is a very useful technique with multiple applications, but often generates multiple undesired products. These may be due to nonspecific annealing of primers at other targets or at similar sequences. To improve the results, changes in the PCR parameters are usually programmed and / or other PCR improved methods performed. Taken into account that the template sequence must be known to choose the adequate primers, some authors [1,2] have found that predigesting the DNA with restriction enzymes that cut outside the region to be amplified, a good amplification of the expected region can be obtained. We have observed a lack of amplification of defined regions depending on the analyzed strains, but with no apparent cause. These amplifications can be taken as *Corresponding author. Tel.: 1 34-93-5811-662; fax: 1 34-93-5812-387. ´ E-mail address: [email protected] (O. Cabre) 0165-022X / 99 / $ – see front matter 1999 Elsevier Science B.V. All rights reserved. PII: S0165-022X( 99 )00017-2
46
E. Baldrich et al. / J. Biochem. Biophys. Methods 40 (1999) 45 – 48
negative results but are, in fact, very poor amplifications behaving as false negatives. We present here a protocol to determine whether a lack of amplification is only a very poor amplification, not detectable by electrophoresis, that might be due to the genomic context. In our particular case we attempted to amplify a 646 base-pair (bp) segment of the white locus of Drosophila melanogaster in some light eye colour mutants appeared in our laboratory and some of its germinal and somatic revertants. This region includes the insertion site of copia retrotransposon characterizing white–apricot allele [3], also considered the zero point of the Levis coordinates [4]. The results obtained by Southern blot showed that the white locus seemed to be intact in all the analyzed mutants, but the region including the empty copia insertion site was not amplified by PCR. The primers were designed from the white locus sequence reported [5,6]. The direct primer (59GGGAATTCATACTTCGTTTTGT-39) anneals at position 1 396 and the reverse primer (59-GTTAGGGAGCCGATAAAGAG-39) anneals at position 2 250, of the Levis coordinates. The standard reactions were carried out in a final volume of 25 ml of reaction mixture containing 50 mM KCl, 10 mM Tris–HCl (pH 8.0), 2 mM MgCl 2 , 0.2 U Taq-polymerase (GIBCO-BRL), 0.15 mM each of dNTPs, 0.25 mM each of primers and 1 ml of DNA template (100–400 ng / ml). Standard PCR parameters were: 948C, 3 min; 29 cycles consisting of: 948C, 45 s; 608C, 45 s; 728C, 90 s; 728C, 5 min. Fig. 1a shows the result of different amplifications carried out are presented here as an example. Fig. 1b shows the Southern blot of the same electrophoresis hybridized with p2k plasmid as probe, to detect poor amplifications and discard nonspecific ones. The p2k plasmid is a 2 kb long BamHI–SacI restriction fragment of the white locus, that contains the sequence to be amplified, cloned in the pTZ-18 plasmid. The expected amplification of 646 bp segment is shown in Fig. 1a, lanes 2 and 4, corresponding to p2k plasmid and wild type strain Canton-S, respectively. Lanes 3 and 8 show controls of no amplification: pTZ-18, which does not contain the sequence to be amplified, (lane 3) and white–apricot mutant, where copia insertion prevents amplification (lane 8). When the amplification was performed with DNA from our mutants no amplification product could be observed, as shown in Fig. 1a, lane 9 for M115 mutant as a example. In order to determine whether the segment was not amplified or only poorly achieved, the PCR products were Southern blotted and hybridized with the p2k probe (Fig. 1b). The expected fragments could then be detected in all the studied mutants as can be seen in lane 9 for M115 mutant. This means that the 646 bp segment was being amplified but with a very low efficiency. Changes in the different amplification reaction conditions or in the mixture components did not improve the result. Amplifications using other primers annealing near the previously described ones (for instance the direct primer 59-ACGAATTGATTTTTAGGGAC-39 that anneals at position 1 745), gave the same negative result. On the other hand, reamplifications and seminested reactions (for instance using the internal direct primer 59-ATGAATTCTGGCAG-39 annealing at 1 385) resulted in a high number of faint bands of unspecific products (Fig. 1a; lane 5: Canton-S as control and lane 10: M115 mutant). In similar cases other authors have proposed a predigestion with restriction enzymes which cut outside the region to be amplified in order to detect viral DNA in human cells
E. Baldrich et al. / J. Biochem. Biophys. Methods 40 (1999) 45 – 48
47
Fig. 1. (a) Example of PCR amplification products obtained after different treatments of templates. Lanes 1 and 14: HaeIII digested fX174 RF DNA length marker. Lane 13: water control. Lanes 2 to 12: 4 ml of final amplification products obtained from the amounts of templates indicated as follows. Lane 2: p2k plasmid (1 ml, | 100 ng / ml). Lane 3: pTZ-18 plasmid (1 ml, | 100 ng / ml). Lane 4: Canton-S (1 ml, | 400 ng / ml). Lane 5: 1 ml of previously amplified DNA from Canton-S. Lane 6: | 400 ng of DNA from Canton-S, digested with BamHI and SacI. Lane 7: 1 ml restriction fragments of | 2 kb in size obtained from Canton-S DNA treated as described in the text. Lane 8: w a (1 ml, | 400 ng / ml). Lane 9: M115 mutant (1 ml, | 400 ng / ml). Lane 10, 1 ml of previously amplified DNA from M115. Lane 11: | 400 ng of DNA from M115, digested with BamHI and SacI. Lane 12: 1 ml restriction fragments of | 2 kb in size obtained from M115 mutant DNA treated as described in the text. (b) The electrophoresis of (a) after Southern blot and hybridization with p2k probe. Detection was carried out by a nonradioactive method (Boehringer).
[1] or human DNA in somatic cell hybrids [2]. We repeatedly found a lack of amplification when the DNA of our mutants and revertants were predigested with restriction enzymes (BamHI and SacI) in order to eliminate unspecific amplifications (Fig. 1a; lane 6: Canton-S as control and lane 11: M115 mutant). Since DNA amplifications of wild type Canton-S strain behaved as expected when used as control, we supposed that some kind of interaction with the genomic context could raise difficulties to the template amplification in our mutants and revertants. For this reason we decided to eliminate, as much as possible, the interferences between our target DNA and the rest of the genome. Therefore we developed a two step protocol. The first step is a digestion of genomic DNA with a restriction enzyme(s) that cuts outside the region to be amplified. The second step is a electrophoresis of digested DNA, followed by recovering the restriction fragments of approximately the desired size by elution from the agarose gel. To achieve this, 5 mg of the total genomic DNA from each sample was digested with BamHI and SacI, generating a 2 kb restriction fragment of the white locus that includes the 646 bp segment to be amplified. The digested DNAs and a DNA length marker were loaded on a 1.5% agarose gel, and run 5 h at 70–80 V. Taking as reference the length of the marker DNA bands, portions of gel containing restriction fragments of approximately 2 kb were cut off from each lane of the gel. The fragments were then eluted in a final
48
E. Baldrich et al. / J. Biochem. Biophys. Methods 40 (1999) 45 – 48
volume of 15–30 ml, using the commercial kit ‘Geneclean’ (Bio101). A 1-ml volume of each sample was used as substrate for PCR. The expected 646 bp fragment was successfully amplified in all the cases when the eluates were used as substrate to carry out the PCR reaction. The amplification was always efficient enough to be directly detected after running 2–4 ml on an agarose gel or, for instance, to be directly sequenced (Fig. 1a, lane 12). We think that the protocol proposed by us in this paper is useful to overcome false negatives in PCR amplification, and improves the methods previously described [1,2]. The previous electrophoretic isolation of target DNA of defined size cut by only one or two enzymes, allows to amplify the target DNA when annoying sequences are unknown and avoids the use of cocktail enzymes.
Acknowledgements ´ y Cultura’ (grant PB96This work was supported by the ‘Ministerio de Educacion 1138, DGES) and E.B. was supported by a doctoral fellowship (FPI) from the ‘Generalitat de Catalunya’.
References [1] Sharma JK, Gopalkrishna V, Das BC. A simple method for elimination of unspecific amplifications in polymerase chain reaction. Nucleic Acids Res 1992;20:6117–8. [2] Hatchwell E. Improving PCR specificity by multiple pre-digestion. Trends Genet 1996;12:460. [3] Bingham PM, Judd BH. A copy of the copia transposable elements very tightly linked to the w a allele at the white locus of D. melanogaster. Cell 1981;5:705–11. [4] Levis R, Bingham P, Rubin G. Physical map of the white locus of Drosophila melanogaster. Proc Natl Acad Sci USA 1982;79:564–8. [5] O’Hare K, Murphy C, Levis R, Rubin GM. DNA sequence of the white locus of Drosophila melanogaster. J Mol Biol 1984;80:437–55. [6] O’Hare K. EMBL / GenBank accession number X02974, 1986.