(http://www.ebiotrade.com/)

GenomiPhi™ DNA Amplification Kit uses a simple and isothermal whole genome amplification method for the representational amplification of human genomic DNA from a small number of cells or limited amount of a precious sample. This method employs the unique biochemical properties of Phi29 DNA polymerase to amplify linear DNA. Microgram quantities of high molecular weight DNA are produced from nanogram amounts of starting material in a simple and robust protocol. Amplified DNA is representative of the original sample, shows little to no amplification bias, and preserves SNP information. No differences were observed when amplified material was substituted for genomic DNA for genotyping, PCR, cloning, and other analyses.

Introduction
Individual sequence differences, or single nucleotide polymorphisms (SNPs), have been implicated in several diseases including Alzheimer's (1), obesity (2), and Crohn's disease (3). The necessary number of SNPs to score for a whole-genome association study may be as high as 500 000 (4). This number of SNPs would need to be scored from each individual in the study. Because even the most sensitive SNP scoring techniques require at least a few nanograms of DNA, availability of genetic material in sufficient quantity and quality to feed these analyses can become a bottleneck.

Amplification of genomic DNA in vitro would be ideal, but it has not met researchers' needs due to limitations in the existing methods. DNA amplification by methods such as primer extension preamplification (5) and degenerate oligonucleotide-primed PCR (6) produce relatively low molecular weight DNA that is not representative of the entire genome. The product is often error prone and unlikely to completely preserve heterozygosity information (7) or correct copying of short repeated elements.

To overcome these limitations, the unique properties of Phi29 DNA polymerase have been exploited to produce an in vitro DNA amplification method that addresses researchers' needs.

GenomiPhi amplification
GenomiPhi DNA Amplification Kit was developed to representatively amplify linear genomic DNA. As illustrated in Figure 1, the amplification method uses bacteriophage Phi29 DNA polymerase to exponentially amplify single- or double-stranded linear DNA templates by strand displacement amplification (8, 9). This method is an extension of the one used in TempliPhi™ DNA Amplification Kit (10, 11) which was optimized for amplification of circular plasmid templates for DNA sequencing.

Fig 1. Schematic diagram of the amplification process with GenomiPhi amplification method. Random hexamer primers anneal to the template DNA at multiple sites. Phi29 DNA polymerase initiates replication at multiple sites on the denatured linear DNA simultaneously. As synthesis proceeds, strand displacement of complementary DNA generates new single-stranded DNA. The subsequent priming and strand displacement replication of this DNA results in the formation of double-stranded DNA.

This method is made possible by the unique properties of Phi29 DNA polymerase. First described by Salas and coworkers (12–14), this enzyme is capable of highly processive DNA synthesis incorporating over 70 000 nucleotides per binding event (12). Thus the random priming and representative synthesis of genomic DNA is unaffected by local extremes of base composition, short tandem repeats, or secondary structure. Processive DNA synthesis minimizes strand switching and panhandle formation.

The second important property of this enzyme is that it supports strand displacement DNA synthesis. The downstream complementary strand is displaced by the enzyme, rendering it single-stranded and available to be primed by random hexamer primers.

Lastly, Phi29 DNA polymerase has an associated 3'-5' exonuclease proofreading activity (13) with a reported error rate of 5 x 10-6 (14), about 100-fold lower than Taq DNA polymerase (15). This produces amplified DNA that has high fidelity to the input DNA.

This isothermal amplification method produces microgram quantities of DNA from nanogram amounts of starting material. Although amplification from single cells is possible and has been demonstrated, the recommended minimum input is 1 ng. This amount represents about 300 copies of human genomic DNA.

One reaction yields about 5 µg of DNA representing a 5000-fold amplification. Although this amount represents about half of the deoxynucleotide triphosphates available for synthesis in the initial reaction components, the DNA polymerase also consumes a significant portion of the triphosphates during the proofreading activity.

DNA amplification with this method is an endpoint reaction. Figure 2 shows a pulsed field gel image of genomic DNA and amplified DNA. The majority of the synthetic products are between 5 and 50 kb in length.

Fig 2. Pulsed field gel analysis. Each lane contains 400 ng DNA. M1 is a 50 kb ladder, M2 is a lambda ladder. G= human genomic DNA, G= amplified DNA. U = undigested sample, D = sample completely digested with EcoR I. Samples were separated in a 1% agarose gel at 65 V for 48 h.

Starting material
The starting material for amplification can be a small amount of genomic DNA prepared by most commercially available DNA purification kits or common in-house methods. For cellular samples such as whole blood, primary cells from clinical samples, or cell culture, a standard alkaline lysis followed by neutralization is sufficient. A minimum input of 1 ng genomic DNA is recommended. Good results have been demonstrated from purified DNA, whole blood, buccal swabs, paper-immobilized blood, and cells from culture. In addition to human genomic DNA, plant chromosomal DNA (purified DNA and from tissue lysates) and microbial genomic DNA have been successfully used in this method.

Applications in PCR
Many genetic analyses use PCR in some form to reduce the complexity of the genome and characterize specific sequences. GenomiPhi amplified material has been used as a substrate for PCR and compared with the parent DNA in simplex, moderate multiplex (6–8-plex) and highly multiplexed (62-plex) PCR configurations. Employing MGB Eclipse™ probes as a real-time assay, the efficiency of PCR was compared. No purification of the GenomiPhi amplified product was performed prior to PCR.

Assays generated against eight different loci on seven different human chromosomes gave equivalent threshold (Ct) values and PCR efficiencies for amplified and genomic DNA, demonstrating that no amplification bias occurred for these loci (Fig 3). Further, the copy number (assuming 1 ng DNA equals 300 copies) served both materials, identically suggesting that quantitation of amplified material is faithful even without purification.

Fig 3. A comparison of real-time PCR performance. An MGB Eclipse probe, designed against beta-actin (7p22.1), was used to assay GenomiPhi amplified DNA (black squares) and non-amplified DNA (blue squares). The amplified DNA was not purified prior to this assay. Real-time PCR was quantitated using an ABI 7900HT Sequence Detection System.

For multiplex PCR, the most robust response is observed when the GenomiPhi amplified product is purified prior to use in PCR or when the reactions were supplemented to enhance efficiency. The high molecular weight product can be purified by precipitation or gel filtration chromatography. The other solution is to add components well known to improve multiplex PCR efficiency, including nonionic detergents, BSA, and additional units of Taq DNA polymerase. Using either of these methods, multiplex PCR performs identically with amplified DNA and genomic DNA.

Applications in SNP genotyping
A useful amplification technology must preserve the original DNA sequence at all positions interrogated on both strands representatively thus preserving heterozygosity. One comparison of the performance of amplified and genomic DNA was performed using the HuSNP™ Mapping Assay (Fig 4). This microarray assay is designed to simultaneously interrogate approximately 1500 SNPs across the entire genome on a single chip. Starting with as little as 1 ng of material in the GenomiPhi amplification reaction gave equivalent results to genomic DNA (approximately 1100 clear calls at greater than 99.3% accuracy; Fig 4). At 10 pg input (approximately 3 copies) to the amplification reaction, lower number of SNP calls and lower accuracy were observed (920 calls, 80.2% accuracy). From this and similar results, a 1 ng minimum input recommendation was established.

Fig 4. SNP scoring on HuSNP microarray. The number of clear calls and accuracy of these calls for CEPH individual 884 DNA is shown. Starting material was either 10 ng of nonamplified genomic DNA, or 30 ng of GenomiPhi amplified DNA amplified from various amounts of starting material.

Several other SNP scoring methods have been employed, each demonstrating that SNP information is preserved when amplified material is used in place of genomic DNA. These methods include MegaBACE™ SNuPe™ Genotyping Kit, an electrophoretic single base extension assay, MGB Eclipse probes, Invader™ genotyping, and direct sequencing of PCR products. Repeat amplification of the amplified product with GenomiPhi DNA Amplification Kit (5 times, serially) revealed that heterozygosity was retained for all SNPs tested (data not shown), suggesting that with this method, critical samples can potentially be replicated as needed.

Applications in microsatellite analysis
High DNA replication fidelity implies that short tandem repeats (STRs) are faithfully and representatively copied. This is the basis of microsatellite genotyping. To assay the fidelity of STR amplification by Phi29 DNA polymerase, amplified and non-amplified DNA from an individual of the CEPH (Centre d'Etudes du Polymorphisme Humaine) reference families (genotype 10859) was genotyped with the 400 markers of the ABI PRISM™ Linkage Mapping set v2.5. Data was analyzed with the MegaBACE 4000 DNA Analysis System and MegaBACE Genetic Profiler Software Suite v2.0. Amplification reactions were carried out with 100 ng, 1 ng, 0.1 ng, or 0.02 ng of reference DNA and each sample was tested in triplicate. An STR sample was categorized as showing loss of allele when one allele was present (homozygote) instead of the expected two (heterozygote). Results were compared against reference STR types given by the manufacturer.

For the control reaction (non-amplified reference DNA), there were seven samples out of the 1200 tests (400 markers in triplicate) that showed loss of allele. For the amplified material, loss of allele was observed in a total of six samples (100 ng input), five samples (1 ng input), three samples (0.1 ng input), or three samples (0.02 ng input). Total scored calls were similar in all cases. Because of the large number of samples, the loss of allele numbers were judged to be very low and functionally identical.

Conclusions
GenomiPhi DNA Amplification Kit provides a reliable method to produce microgram quantities of DNA from nanogram amounts of genomic DNA. The amplified DNA is high molecular weight, high fidelity, and representative of the input material. Sequence fidelity and utility has been demonstrated in PCR, SNP and STR applications. Further, because the amplified product can be re-amplified, this method could be used in an analogous way to immortalized cell lines for extensive genomics research.