drop-in, drop-out allele-specific pcr

Drop-In, Drop-Out Allele-Specific PCR 171

MOLECULAR BIOTECHNOLOGY Volume 28, 2004

RESEARCH

171

Molecular Biotechnology © 2004 Humana Press Inc. All rights of any nature whatsoever reserved. 1073–6085/2004/28:3/171–174/$25.00

*Author to whom all correspondence and reprint requests should be addressed. 1Research Service 151, New York Campus, New YorkHarbor Healthcare System, Department of Veterans Affairs, 423 East 23 Street, New York, NY 10010, USA. E-mail:[email protected] ; 2Department of Molecular and Experimental Medicine, The Scripps Research Institute; 3Medical Service, VA Boston HealthcareSystem and Boston University School of Medicine.

Drop-In, Drop-Out Allele-Specific PCR

A Highly Sensitive, Single-Tube Method

Alice Alexander,1 Nivedita Subramanian,1

Joel N. Buxbaum,2 and Daniel R. Jacobson1,3,*

AbstractAllelotyping large numbers of samples by allele-specific polymerase chain reaction (PCR) can be prob-

lematic if the DNA samples to be tested are of highly variable concentration. On the one hand, analysis ofdilute DNA samples often requires nested PCR to produce a product of sufficient yield to be detectable onethidium bromide-stained agarose gels. Such two-step assays require additional reagents, are labor-inten-sive, and have a higher risk of contamination. On the other hand, the specificity of allele-specific PCRassays can be lost at high input DNA concentrations. Large population-based genetic studies using DNAfrom varied sources would benefit from one-tube assays that could detect mutations in samples over a widerange of concentration. We describe a one-tube nested allele-specific PCR-based assay, in which the inputDNA concentration has little effect on the assay’s yield or specificity. An assay using this method is highlysensitive and specific, and was used to type several thousand DNA samples, obtained from various sources,for a G to A transition at human transthyretin codon 122. Similar assays could be readily adapted to anyhigh-throughput allelotype assay where input DNA is of highly variable concentration.

Index Entries: Allele-specific PCR; nested PCR; drop-in, drop-out PCR; mutation frequency; allelefrequency; transthyretin.

1. IntroductionWhen DNA samples from various sources are

assayed in large studies of allele frequencies, theDNA concentration may vary widely amongsamples, presenting technical problems for poly-merase chain reaction (PCR)-based analysis, es-pecially allele-specific PCR. Highly concentratedDNA samples cause allele-specific PCR to lose itsspecificity (1), whereas very dilute DNA samplesoften require two-step “nested” PCR reactions.Nested PCR is both labor-intensive and moresubject to contamination than one-step PCR.

To address these issues, we adapted the “drop-in, drop-out” PCR method (2–4), for use with al-lele-specific PCR. The goal of previouslydescribed drop-in, drop-out assays has been to

maximize PCR product from small amounts ofstarting material. Our goal, in contrast, is to en-able allelotyping on samples of widely varyingDNA concentration. Drop-in, drop-out PCR hasapparently not been previously described with al-lele-specific PCR, although these two methodsseem particularly suitable for use together.

Our protocol is a highly sensitive, one-tubeassay containing both the primary and nested prim-ers, and it maintains allele specificity independentof DNA sample concentration. It compensates forhigh or low amounts of input DNA by using a two-phase PCR cycling program and limited concen-trations of outer primers. In the primary cyclingphase, at a high annealing temperature, only theouter primers hybridize to the template DNA.

172 Alexander et al.


When the outer primers are depleted they “drop-out” of the reaction. In the secondary phase, at alower annealing temperature, the nested primers,including the allele-specific primer, “drop-in” tothe reaction and hybridize to the DNA synthe-sized in the primary phase.

The method is particularly suitable in assaysscreening for relatively rare alleles, where verylarge numbers of DNA samples must be tested. Inthese situations, the less labor-intensive, one-tubemethod serves as the first screen to identify therare samples containing mutations. Mutation-positive samples can then be confirmed using amethod based on restriction enzyme analysis (5).Such methods confirm the mutation and also de-termine whether the sample is heterozygous orhomozygous, which assays based on allele-spe-cific PCR cannot do.

We have used this protocol successfully toscreen more than 4000 African-American genomicDNA samples for a G to A transition encoding thetransthyretin V122I mutation. Previous studieshave reported that the allele prevalence is 0.019 inAfrican-Americans and is associated with cardiacamyloidosis (5–7).

2. Materials and MethodsGenomic DNA, in varying concentrations, was

obtained from more than a dozen sources. Con-trol DNA samples were previously identified asencoding the normal or V122I transthyretin vari-ant sequence, using PCR and restriction analysis(5).

Each PCR reaction mixture (25 µL total vol-ume) contained 1.2 mM MgCl2, 40 µM eachdNTP, and 1.25 units of Taq DNA polymerase per25 µL reaction. Cycling consisted of: phase I: 25cycles of 94°C for 10 s, 70°C for 20 s, and 74°Cfor 90 s; phase II: 30 cycles of 93°C for 5 s, 53°Cfor 15 s, and 74°C for 30 s; phase III: 3-min elon-gation at 74°C. Of each reaction, 4.5 µL wasloaded on a 2.5% agarose gel and stained withethidium bromide.

Primers were added to reactions in concentra-tions determined empirically (Table 1). Thelength and GC content of the primers were cho-sen so that in the 25 cycles comprising phase I,

only the long, GC-rich “primary” primers, presentat low concentration, could bind to the genomicDNA under stringent conditions (annealing tem-perature 70°C). In the following 30 cycles, at alower annealing temperature (53°C), the short,AT-rich nested primers (including the allele-spe-cific primer) prime the synthesis of PCR productsthat are visible on ethdium bromide-stained agar-ose gels (Fig. 1). Primer 1 is present at too low aconcentration (0.3 pmol per 25-µL reaction) forthe primary 544-bp product to be routinely vis-ible on ethidium bromide/agarose gels. Primer 2is present in a threefold molar excess over primer1 because primer 2 also participates in phase II ofthe reaction.

In phase II, the low annealing temperature al-lows the short primers, 3 and 4, to hybridize tothe 544-bp primary PCR product. The allele spe-cific primer (primer 4) forms a perfect match onlywith DNA containing the V122I mutation. If noV122I DNA is present, the DNA mismatches withthe 3'-end of primer 4, and no PCR product is syn-thesized. The assay conditions include low MgCl2and nucleotide concentrations. These conditionsprevent extension of a mismatched base at the 3'-end of the allele-specific primer (referred to asbreakthrough amplification of normal DNA).Thus, the allele-specific primer allows DNA syn-thesis to occur only in the presence of the mutantallele.

Primers 3 and 2 prime a seminested 291-bpproduct, synthesized from either normal or mu-tant DNA; this band serves as a positive controlfor the PCR reaction. Primers 3 and 4 lead to syn-thesis of a 238-bp product only if the starting ma-terial contains the transthyretin V122I mutation.Thus, the 291-bp product appears in all PCR re-actions, whereas the 238-bp product is only pro-duced in the presence of the mutation (Fig. 1).

3. Results and DiscussionThe validity of the one-tube nested allele-spe-

cific assay was confirmed by amplification of hu-man DNA samples of known codon 122 allelestatus. Ten control samples known to contain themutation and 16 samples containing only the nor-mal allele were scored correctly using the assay

Drop-In, Drop-Out Allele-Specific PCR 173


Fig. 1. Assay concept and results. The primary and nested PCR products are as shown. The position on the gelof the primary 544-bp product is indicated, but this band is usually not visualized because of the low concentra-tion of primer 1. The 291-bp product often appears as a double band on agarose gels, as seen here. Because primer3 is present in 10 times the concentration of primer 2, it is likely that some single-stranded product is synthesized,which may have a faster mobility due to conformational factors. The 291-bp band is often not seen as a doubleband when the substrate is mutant DNA (Figs. 2 and 3). This may be due to a lesser concentration asymmetrybetween primer 3 and 2, because, when mutant DNA is the template, primer 3 will also prime with primer 4 in thesynthesis of the 238-bp allele-specific band.

Table 1Primer Specifications

Primers Amount per Sequence GC Predicted25-µL reaction content Tma

Primary Primer 1 0.3 pmol 5'-CTGTGGTTGGCAGCCACTATTGCAGCAGCTC-3' 58% 83.5Primer 2 1.0 pmol 5'-CATGAAATCCCATCCCTCGTCCTTCAGGTCC-3' 55% 79.6

Secondary Primer 3 10.0 pmol 5'-CTGTTCAAACTGTTCCAAA-3' 37% 56.1Primer 4 30.0 pmol 5'-CTTGGGATTGGTGAT-3' 47% 51.1

aMelting temperature of oligonucleotide, as reported by the manufacturer (Sigma Genosys or DNA International).

(data not shown). In addition, all new samplesscored as mutation-positive in the present assaywere confirmed by restriction analysis (5). By re-serving the more time-intensive restriction analy-sis to the samples scored as positive in theallele-specific assay, we limit the number of re-striction analyses to the relatively small percent-age carrying the variant allele. More than 4000African-American genomic DNA samples werescreened for a G to A transition encoding thetransthyretin V122I mutation. The percentage ofTTR Ile122-positive samples was similar to the

previously determined population allele preva-lence of 0.019 (Jacobson, D.R., et al., manuscriptin preparation).

To show that sensitivity was improved by in-corporating nested PCR, our one-tube assay wascompared to an assay in which primer 1 was miss-ing, eliminating the primary phase of the reaction(Fig. 2). The three-primer assay was 100-fold lesssensitive than the four-primer assay.

Limiting the concentration of primer 1 has theimportant effect of limiting the amount of the 544-bp DNA product, which is the template for the

174 Alexander et al.


allele-specific PCR in phase II. To demonstratethe accuracy of the assay over widely varied in-put DNA concentrations, serial dilutions of onenormal and one heterozygous variant DNA wereassayed; the assay yields correct results over fourorders of magnitude of input DNA concentration(Fig. 3). In the normal lanes, only one band wasvisible, with no “breakthrough amplification” ofthe variant band at any concentration of startingmaterial. Thus, limiting the 5' primary phaseprimer to 0.3 pmol in a 25-µL reaction places anupper limit on the amount of primary PCR prod-uct available for the nested reaction, preventingloss of assay specificity, even when starting witha high input DNA concentration.

AcknowledgmentsThese studies were supported by an Established

Investigator Award of the American Heart Asso-ciation (Dr. Jacobson) and Merit Review Fundsfrom the Department of Veterans Affairs (Dr.Jacobson), and NIH R01 AG15916 (Dr. Buxbaum).

Fig. 3. Serial dilutions of starting DNA. The assayis shown using serial 1:10 dilutions of DNA contain-ing either the normal or heterozygous transthyretinV122I-containing DNA. The 291-bp band is ampli-fied in reactions containing both normal or mutantDNA. The 238-bp allele-specific band, seen only inthe reactions containing mutant DNA, is detected atall dilutions. There is no “breakthrough amplification”of the 238-bp band in reactions containing only nor-mal DNA. Thus, allele specificity is maintained whenDNA is diluted over four orders of magnitude.

Fig. 2. A comparison of the assay with and withoutprimer 1. The DNA used in all reactions was a V122Iheterozygote. For cycling conditions A, samples werecycled as described in Subheading 2. For cycling con-ditions B, samples were cycled as in A, except that theannealing temperature of the primary phase of the re-action was changed to 53°C. All other conditions forthe two assays were identical.

References1. Sarkar, G., Cassady, J., Bottema, C., and Sommer, S.

(1990) Characterization of polymerase chain reaction am-plification of specific alleles. Anal. Biochem. 186, 64–68.

2. Erlich, H. A., Gelfand, D., and Sninsky, J.J. (1991)Recent advances in the polymerase chain reaction.Science 252, 1643–1651.

3. Pecharatana, S., Pickett, M. A., Watt, P. J., and Ward,M. E. (1993) Genotyping ocular strains of Chlamydiatrachomatis by single-tube nested PCR. PCR Meth-ods Appl. 3, 200–204.

4. Tilston, P. and Corbitt, G. (1995) A single tube nestedPCR for the detection of hepatitis C virus RNA. J.Virol. Methods 53, 121–129.

5. Jacobson, D. R. (1992) A specific test for transthyretin122 (Val–Ile) based on PCR-primer introduced restric-tion analysis (PCR-PIRA): confirmation the gene fre-quency in Blacks. Am. J. Hum. Genet. 50, 195–198.

6. Jacobson, D. R., Pastore, R., Pool, S., et al. (1996)Revised transthyretin Ile 122 allele prevalence in Af-rican-Americans. Hum. Genet. 98, 236–238.

7. Jacobson, D. R., Pastore, R. D., Yaghoubian, R., etal. (1997) Variant-sequence transthyretin (isoleucine122) in late-onset cardiac amyloidosis in blackAmericans. N. Engl. J. Med. 336, 466–473.

drop-in, drop-out allele-specific pcr

Documents