|
Development of a Multiplex Ligase Detection Reaction DNA Typing AssayPhillip Belgrader1, Francis Barany2 and Matthew Lubin2
Current methods for DNA typing at our facilities utilize the polymerase chain reaction (PCR) for typing single nucleotide and tandem repeat polymorphisms. Unfortunately, these PCR-based methods are relatively expensive, time consuming and are not well suited for total automation. The ligase detection reaction (LDR) when used in conjunction with PCR offers distinct advantages. Taq DNA ligase does not ligate mismatches at the junction site of two adjacent primers and, since it is thermostable, permits several cycles of ligation to occur. In addition, it is feasible to integrate DNA purification, PCR, and LDR into a completely automated system for high throughput typing. A variety of detection platforms (i.e. plate reader, capillary electrophoresis system, mass spectrometer, or microchip) may be implemented for analysis of the LDR product. We have developed a coupled multiplex PCR-LDR assay to type single point variations at 12 biallelic loci. The 12 loci are PCR amplified in a single reaction using a unique method that produces similar amounts of multiplexed products without the need to carefully adjust primer concentrations or PCR conditions. These PCR products are used in a single LDR to generate LDR products which are resolved and typed on an Applied Biosystems 373 DNA Sequencer to generate an LDR profile. Our ability to easily generate similar amounts of product in a 12-locus multiplex PCR amplification is the basis for expanding the assay in order to type 30 biallelic loci, providing a power of discrimination of at least 1 individual in 1 billion. Single nucleotide variations are quite common in mammalian genomes and exist at a frequency of about one per four hundred base pairs (1). These types of polymorphisms are being exploited by the medical and forensic fields to diagnose genetic diseases and identify crime suspects or victims (2,3). In order to minimize the time to process samples, increase the sensitivity of detection, increase the number of loci typed, and work with DNA that is not pristine, several PCR-based methods have been developed that are routinely used to discriminate single base differences. These include allele-specific amplification (4), allele-specific and reverse oligonucleotide hybridization (5,6), genetic bit analysis (GBA) (7), the oligonucleotide ligation assay (OLA) (8), and the ligase detection reaction (LDR) (9). The OLA and LDR take advantage of DNA ligase which has been demonstrated to be ideal for discriminating point variations in a multiplexed assay (10-12). An equimolar mixture of two detecting oligonucleotides and one common oligonucleotide (see Figure 3) is hybridized to denatured PCR amplified target. The detecting or allele-specific oligonucleotides anneal immediately adjacent to the 5' end of the common oligonucleotide, resulting in the formation of a short DNA duplex containing a nick at the junction site between the oligonucleotides. The two allele-specific oligonucleotides are in competition for hybridization to the denatured target, and depending upon which of the detecting oligonucleotides has hybridized, the 3' end of the allele-specific oligonucleotide will have perfect complementarity (match) or will contain a single base mismatch. If there is a match, then the junction between the allele-specific and common probes will be covalently linked by DNA ligase to generate a uniquely sized ligation product. A single base mismatch at the junction inhibits ligation, and thus point variations are distinguished (9,13). LDR has the distinct advantage over OLA in that it makes use of the thermostable Taq DNA ligase (9,12). Taq DNA ligase dramatically improves the sensitivity of the reaction since the ligations may be repeated to linearly increase product. In addition, the ligation reactions can be performed at higher temperatures to increase annealing stringency, thereby increasing specificity. This report describes the use of a coupled multiplex PCR-LDR assay for forensic identification. Over 80 gene regions with single base polymorphisms were identified from the Human Genome Database. Twelve of these (Table 1) were amplified and typed in a single PCR-LDR assay. Our ability to easily type 12 loci is the basis for increasing the number of sites typed in a single assay to 30. DNA Purification and Oligonucleotide Synthesis DNA isolation was performed from whole blood using the Purgene DNA Isolation Kit (Gentra Systems, Inc., Minneapolis, MI). Oligonucleotides were assembled by standard phosphoramidite chemistry on an Expedite DNA synthesizer (Perseptive Biosystems, Framingham, MA). Oligonucleotides 5'-end labeled with 6-FAM, TET, and HEX were synthesized using the appropriate dye phosphoramidites (Perkin Elmer-Applied Biosystems, Foster City, CA) and purified with Oligonucleotide Purification Cartridges (Perkin Elmer-Applied Biosystems) following the manufacturer's protocol (14). LDR common oligonucleotides were phosphorylated at the 5' end either during the synthesis with Phosphate-ON (Clontech Laboratories, Palo Alto, CA) or post-synthesis, using T4 polynucleotide kinase (Boehringer Mannheim, Indianopolis, IN). Twelve amplification products were generated in a single PCR (Figure 1). A volume of 25 ml of PCR buffer (10 mM Tris/HCl pH 8.3, 10 mM KCl, 4 mM MgCl2, 200 mM dNTPs), 10-100 ng of genomic DNA, PCR hybrid primer pairs 1-12 (2 pmol of each primer), and 1.3 units of AmpliTaq DNA polymerase Stoffel fragment (Applied Biosystems-Perkin Elmer) was placed in a MicroAmp reaction tube (Applied Biosystems-Perkin Elmer). The hybrid primers were designed to have gene-specific 3' ends and 5' ends corresponding to one of two pairs of PCR universal or "zip code" primers. Amplification was attained by thermal cycling for 1 cycle of 96ºC for 15 sec to denature, then 15 cycles of 94ºC for 15 sec to denature and 65ºC for 60 sec to anneal and extend. An equal volume of PCR buffer containing PCR zip code primers (25 pmol of each primer) and 1.3 units of AmpliTaq DNA polymerase Stoffel fragment was added to the tube, and thermal cycling proceeded for another 25 cycles with the annealing temperature lowered to 55ºC. All thermal cycling was achieved with a GeneAmp PCR System 9600 thermal cycler (Applied Biosystems-Perkin Elmer). Multiplex Ligase Detection Reaction (LDR) A 5 µl aliquot of PCR product was diluted in 20 µl of LDR mix containing 50 mM Tris/HCl pH 8.5, 50 mM KCl, 10 mM MgCl2, 1 mM NAD+, 10 mM DTT, LDR oligonucleotide sets 1-12 (200 pmol of each primer; Figure 3), and 10 units of Thermus aquaticus DNA ligase. Thermal cycling was performed for 1 cycle of 95ºC for 2 min to denature, then 20 cycles of 95ºC for 30 sec to denature and 65ºC for 4 min to ligate. Gel Electrophoresis and Fluorescence Detection A 3 µl aliquot of PCR or LDR sample was mixed with 3 µl of formamide containing fluorescently labeled Genescan-2500 (TAMRA) size standard (Applied Biosystems-Perkin Elmer). The sample was heated at 95ºC for 2 min, quick cooled in ice, and electrophoresed through a denaturing 8 or 10% polyacrylamide gel in an Applied Biosystems 373 DNA Sequencer running Genescan version 1.2 software. The sizes of the fluorescently labeled products were automatically computed by the Genescan analysis software using the local Southern method. The combined use of PCR for amplification and LDR for discrimination of alleles is a powerful approach for multiplexed DNA typing. PCR enables high sensitivity by generating a substantial amount of product for detection; LDR provides the high discriminatory power of a thermostable ligase to distinguish point variations. In the multiplex PCR-LDR assay we describe for identity testing, 12 amplicons derived from 12 different biallelic loci are generated in a single PCR and the alleles present for each locus are identified in a single LDR. Developing a multiplex PCR that yields equivalent amounts of each PCR product can be difficult and laborious. This is primarily due to the varying annealing rates of the primers in the reaction. Typically, primer concentrations, salt concentrations, and annealing temperatures are adjusted in an effort to balance the annealing rates of all the primers in the reaction. Unfortunately, as the number of amplicons in a PCR is increased, it becomes more difficult, if impossible, to work out conditions to obtain an equal amount of each product. The success of our PCR-LDR typing assay was the result of a new strategy to perform multiplexed PCR (Figure 1). Twelve gene regions were chosen for typing based on information available in the Human Genome Database (Table 1). Each region was well characterized and harbored a single point variation in which only two alleles were known to exist. Fifteen cycles of PCR amplification were performed using 12 pairs of hybrid primers (numbered according to their respective loci in Table 1). Each hybrid primer consisted of a gene-specific 3' region (16-29 bases) and a 5' region (22 bases) corresponding to one of two sets of zip code primers. Forward and reverse hybrid primers for loci 1,3,5,7,10 and 12 contained 5' end regions identical to zip code primers ALg1 and BLg2 respectively. Forward and reverse primers to loci 2,4,6,8,9 and 11 contained 5' end regions identical to zip code primers CLg3 and DLg4 respectively. The low concentration of the hybrid primers in the reaction was rate limiting, causing the primers to be virtually exhausted during the reaction. This allowed products with low amplification efficiencies to "catch up with" those that had high amplification efficiencies. At the completion of the 15 cycles, a high concentration of the two pairs of zip code primers were added. The upstream zip code primers ALg1 and CLg3 were fluorescently labeled with 6-FAM and TET respectively. Thermal cycling continued for another 25 cycles, and each product was now amplified with the zip code primers. The PCR products were separated on an Applied Biosystems 373 DNA Sequencer and the electropherograms clearly showed 12 distinct products (Figure 2). The uniform amount of each product was attributed to the similar size of each amplicon and the use of the zip code primers, which annealed with identical affinities to the 12 amplicons without the need to carefully adjust reaction conditions. The computed sizes of the products, which ranged from 135 to 175 bp, matched exactly to their actual sizes (Table 2). It should be noted that the multiplex PCR could have been designed to utilize one zip code primer pair instead of two, and all the products would have been labeled with the same fluorescent dye. However, the dual labeling approach (Table 2) made it much easier to distinguish each product on the electropherograms (Figure 2, compare panel A with panels B and C). Twelve sets of oligonucleotides were designed and synthesized. Each set consisted of two allele-specific oligonucleotides and a common oligonucleotide (Figure 3). Each pair of discriminating allele-specific oligonucleotides in LDR oligonucleotide sets 1,2,3,4,6,7,8 and 9 were the same size, with one oligonucleotide labeled with 6-FAM and the other labeled with TET (Table 2 and Figure 3). For LDR oligonucleotide sets 5, 10, 11 and 12, each pair of allele-specific oligonucleotides differed by 2 bases (the larger oligonucleotide had a 5' tail that was not complementary to the target sequence), and both oligonucleotides were labeled with HEX. To avoid any possibilities of labeled PCR product interfering with the detection of ligation products, PCR product amplified using unlabeled PCR zip code primers served as the target for the LDR. The multiplexed LDR was performed using all 12 oligonucleotide sets in a single reaction (Figure 3). Detection and analysis was achieved using the 373 DNA Sequencer and LDR profiles of two individuals are shown in Figure 4. When each fluorescent dye is analyzed independently (Figure 4, panels B-D and F-H), it is very easy to determine the alleles present for each locus. A simple "A" or "B" code was assigned to each LDR product (Table 2) and used to score the genotypes (Table 3). The first individual (Figure 4A-D; Table 3, individual 1) was heterozygous at polymorphic sites 1-7 and 10-12. Heterozygosity at sites 1-4 and 6-7 was indicated by the detection of both 6-FAM and TET labeled products (Figure 4, panels B and C) at the respective positions on the electropherograms. Heterozygosity at sites 5 and 11-12 was indicated by the presence of two HEX labeled products, differing in size by 2 bases, for each of these loci (Figure 4, panel D). In contrast, the one product detected at sites 8 and 9 (Figure 4, panels B and C) established that each of these loci was homozygous. The second individual (Figure 4E-F; Table 3, individual 2) was heterozygous only at sites 3,5,6 and 9 and homozygous at sites 1,2,4,7,8 and 10-12. There was a total of 8 differences in the genotypes at these positions between the two persons. Three additional individuals were typed, and all 5 persons were determined to each have distinct genotypes based on the 12 loci (Table 3). The multiplex PCR-LDR typing of 12 biallelic loci for 5 individuals exquisitely demonstrates the fidelity and robustness of this technique. Future work involves dramatically increasing the number of loci typed in the assay and implementing the entire PCR-LDR process into a totally automated system. This system could include a plate reader to analyze the LDR products; however, the small size of the LDR products (Table 3) makes it feasible to resolve and detect them on other platforms such as capillary electrophoresis instruments (11) and matrix-assisted laser desorption time-of-flight mass spectrometers. Ideally, the entire PCR-LDR assay including detection would be implemented into a self-contained microchip, enabling faster analysis times, less reagents, higher reaction efficiencies, and no risk of exposure to an open environment. The opinions or assertions herein are those of the authors and do not necessarily reflect the views of the Department of the Army or of the Department of Defense.
Table 1. List of Polymorphic Sites Analyzed
Note: The site numbers are specific point variations located within the respective genes. Het. = heterozygosity. Table 2. List of PCR and LDR Products
Table 3. Genotypes Determined by PCR-LDR for 5 Individuals.
Figure 1. Scheme for the multiplex PCR amplification of 12 loci using zip code primers.
Hybrid PCR primers were designed to have gene-specific 3' ends (black boxes) and 5' ends (black boxes) corresponding to one of two sets of zip code primers (black boxes). All 24 hybrid primers were used at low concentration in a 15 cycles PCR. After this the two pairs of zip code primers were added at higher concentrations and the PCR was continued for an additional 25 cycles. The upstream zip code primers were synthesized with either FAM or TET fluorescent labels. Figure 2. Electropherogram results of the multiplex PCR.
The products were separated on a 373 DNA Sequencer. Panel A shows the electropherograms for the FAM- and TET-labeled products combined. Panel B shows the FAM-labeled products alone. Panel C shows the TET-labled products alone. The use of the zip code system produces similar amounts of multiplexed products without the need to carefully adjust primer concentrations or PCR conditions. Figure 3. Diagram of the 12 sets of LDR primers.
LDR primer sets were designed in two ways: (i) allele-specific primers were of the same length but contained either FAM or TET label; or (ii) the allele-specific primers were both labeled with HEX but differed in length by two bases. Figure 4. PCR-LDR of 12 biallelic genes for human identification.
PCR of the 12 polymorphic genes was performed as described in Fig. 1, however, zip code primers were not fluorescently labeled. PCR products were diluted into a ligase buffer containing 36 LDR primers (see Figure 3). After 20 cycles of LDR, the ligated products were resolved on a 373 DNA Sequencer. Panel A and E show the 12 locus PCR/LDR profiles of two individuals. Panels B, C, and D show respectively the FAM, TET, and HEX data for the individual in Panel A. Panels F, G, and H show respectively the FAM, TET, and HEX data for the individual in panel E. The individual in panel A is homozygous only at locus 6 (ALDOB) and locus 8 (IGF). The individual in panel E is heterozygous only at loci 3 (C6), 5 (NF1), 6 (ALDOB) and 8 (IGF). This demonstrates that PCR-LDR can simultaneously distinguish both homozygous and heterozygous genotypes at multiple positions. Go to proceedings home page
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
