Whole Genome Amplification (WGA)
Protocols & Applications

PCR-based WGA

There are 3 main PCR-based WGA techniques. These are degenerate oligonucleotide PCR (DOP-PCR) (1), primer extension preamplification (PEP) (2) or derivatives thereof, and adaptor-ligation PCR (3). The main difference between the techniques is that PEP uses a preamplification step to add primer binding sites to small DNA fragments for later WGA by PCR, while adaptor-ligation PCR uses adaptors ligated to small DNA fragments to create PCR primer binding sites. PEP utilizes random primers and a low PCR annealing temperature. Less frequently used today, DOP-PCR uses semi-degenerate oligonucleotides (i.e., CGACTCGAGNNNNNNATGTGG) and an increasing annealing temperature. The use of Taq DNA polymerase in both techniques limits the fragment lengths to 3 kb (average fragment sizes are 400–500 bp) and also introduces a number of errors into the sequence. Furthermore, these techniques have been found to exhibit incomplete genome coverage and amplification bias – where a sequence is overrepresented in the amplified DNA due to preferential binding of the primers to specific regions.

Back to top

Multiple displacement amplification WGA

Multiple displacement amplification (MDA) involves the binding of random hexamers to denatured DNA followed by strand displacement synthesis at a constant temperature using the enzyme Phi29 polymerase. Additional priming events can occur on each displaced strand leading to a network of branched DNA structures (see figure Schematic representation of MDA).

Phi29 polymerase does not dissociate from the genomic DNA template allowing the generation of DNA fragments up to 100 kb without sequence bias. The enzyme has a 3'– 5' exonuclease proofreading activity and provides error rates up to1000 times lower than Taq DNA polymerase-based methods (see PCR-based WGA).

PCR-based WGA methods (e.g., PEP and adaptor-ligation PCR) are affected by secondary DNA structures. These structures can cause enzyme slippage (4) or dissociation of the enzyme from the template resulting in nonspecific amplification artifacts, incomplete coverage of loci, and short DNA fragments (less than 1 kb) that cannot be used in many downstream applications (5). In contrast, the strand-displacing enzyme Phi29 polymerase resolves secondary DNA structures (see figure Secondary DNA structures), enabling accurate and uniform amplification of the genome. The long DNA fragment lengths generated using the highly processive Phi29 polymerase ensure that Phi29-amplified DNA covers the whole genome, enabling consistent and unbiased locus representation.

Unbiased amplification

The long DNA fragment lengths generated using the highly processive Phi29 polymerase ensure that MDA-amplified DNA covers the whole genome, enabling consistent and unbiased locus representation.

In a bias-study, various amounts of genomic DNA (0.3–300 ng) were amplified by using MDA, DOP-PCR, and PEP. The relative representation of 8 loci was determined using quantitative real-time PCR (see figure Highly representative amplification using MDA). Locus representation was determined by comparison to 1 µg of unamplified control DNA (5).

Due to unequal amplification of different loci caused by unresolved secondary structures, PCR-based WGA methods like DOP-PCR or PEP exhibit frequent locus dropout. MDA shows highly representative DNA amplification and minimal risk of locus dropout.

The importance of DNA denaturation in WGA

Long stretches of intact DNA are the ideal template for successful WGA to ensure the highest sensitivity of downstream assays. Assays performed on amplified, fragmented DNA are not as sensitive or reliable, as the risk of a DNA breakpoint in the locus of interest is higher (see figure Effect of DNA fragmentation on WGA). For optimal results, it is important that the template DNA is completely denatured. However, heat denaturation at 95°C can damage the template DNA, resulting in incomplete and less consistent locus representation. Alkaline DNA denaturation circumvents these problems and enables uniform DNA amplification across the whole genome with minimal sequence bias.

In addition to affecting locus representation, DNA denaturation at 95°C should be avoided, as it also affects the length of DNA products amplified by MDA by prematurely dissociating Phi29 from the DNA template. In general, Phi29 polymerase is capable of replicating up to 100 kb without dissociating from the genomic DNA template. DNA amplified using MDA has an average product length greater than 10 kb, with a range between 2 kb and 100 kb. Thus, the DNA yielded by MDA amplification is ideal for use in downstream applications that require long DNA fragments (e.g., RFLP analysis and Southern blotting).

WGA from single cells

In order to make sure that the whole genome is used for single-cell WGA, the whole intact cell is added to the WGA reaction. Because every DNA break results in the loss of the sequence information at that site, MDA is a highly suited for amplifying the whole genome. The reaction is used for the WGA since Phi29 polymerase is capable of replicating up to 100 kb without dissociating from the genomic DNA template.

In a study, the genomic DNA of single cells (3 replicates) was amplified using MDA. Approximately 40 µg of DNA was generated during single-cell WGA. Single-cell WGA DNA was compared with WGA DNA from 1000 cells and nonamplified DNA in whole genome sequencing. Whole genome sequencing was performed on the Illumina MiSeq instrument starting with 2 µg of DNA (WGA-DNA or nonamplified genomic DNA). Comparable sequence coverage was observed for gDNA and single-cell amplified DNA. Comparison of nonamplified and REPLI-g amplified DNA revealed error rates in a similar, very low, percentage range (see figure Comparable NGS results using genomic or single-cell amplified DNA).

Working with fragmented DNA

MDA requires average genomic DNA fragment sizes of approximately 2 kb in order to amplify DNA without introducing any bias. Fragmented or low-quality DNA can be used as long some DNA fragments are above 2 kb in length; although fragments can be ligated to create longer DNA. This is because randomly fragmented DNA should contain multiple intact copes of each locus. However, to ensure accurate locus representation, the starting amount of template DNA should be increased.

Working with fixed tissue samples: DNA from FFPE tissue

Formalin fixation is a commonly used technique for preserving tissue samples for paraffin embedding (FFPE). The fixation ensures the preservation of tissue architecture and cell morphology by cross-linking biomolecules. Different sample types may require a different fixation procedure: tissues with a soft consistency, such as breast tissue samples, usually require a longer fixation step to preserve tissue morphology. Longer fixation times may result in two effects:

• A higher degree of cross-links is generated between biomolecules
• A higher degree of DNA fragmentation occurs — resulting in small DNA fragments of usually several hundred base pairs in length

Fragmentation of DNA causes a reduction of genome equivalents that can be detected by PCR and therefore has a large effect on downstream assays (see Critical factors that influence PCR-based analysis following MDA).

Although no simple method is currently available to determine the degree of cross-linking within a sample, gel electrophoresis of DNA isolated from FFPE samples can give valuable hints about the quality of the DNA.

Back to top

Critical factors that influence downstream PCR-based analysis following MDA

Detection of a specific genetic loci from a paraffin-embedded tissue sample using PCR is strongly influenced by the following factors:

• Copy number
• The degree of cross-linking
• Amplicon size

Copy number: The larger the amount of DNA, and therefore the copy number of the genome, the more likely a specific locus will be detected by PCR after whole genome amplification. However, quantification of DNA from paraffin-embedded samples can be challenging due to the varying amount of contaminating substances isolated with the DNA. Contamination can be identified using a UV scan. In contrast, using single A₂₆₀ measurements instead of a UV scan (220 to 320 nm) will lead to an overestimation of DNA concentration. By overestimating the amount of input DNA in PCR or other downstream experiments, low performance may be mistakenly observed.

Cross-links: The higher degree of crosslinking in a DNA sample, the lower the performance of an amplification reaction such as whole genome amplification.

Amplicon size: The smaller the size of the amplicon used for PCR-based analysis of FFPE DNA, the greater the chance of detecting a specific locus. Real-time PCR of intact DNA compared with DNA from paraffin-embedded samples shows that increasing amplicon sizes dramatically reduces the number of detectable genome equivalents.

Back to top

WGA applications

DNA amplified using MDA-based WGA is suitable for use in:

• Next-generation sequencing
• Genotyping using microarrays
• Comparative genome hybridization studies (CGH)
• Single nucleotide polymorphism (SNP) genotyping
• Sanger sequencing
• STR/microsatellite analysis
• Haplotyping

Back to top

References

1. Telenius, H., Carter, N.P., Bebb, C.E., et al. (1992) Degenerate oligonucleotide-primed PCR: general amplification of target DNA by a single degenerate primer. Genomics 13, 718.
2. Zhang, L., Cui, X., Schmitt, K., et al. (1992) Whole genome amplification from a single cell: Implications for genetic analysis. Proc. Natl. Acad. Sci. USA 89, 5847.
3. Rosenthal, A., and Jones, D.S. (1990) Genomic walking and sequencing by oligo-cassette mediated polymerase chain reaction. Nucleic Acids Res. 18, 3095.
4. Viguera, E., Canceill, D., Ehrlich, S.D. (2001) Replication slippage involves DNA polymerase pausing and dissociation. EMBO J. 20, 2587
5. Dean, F.B. et al. (2002) Comprehensive human genome amplification using multiple displacement amplification. Proc. Natl. Acad. Sci. USA 99, 5261.

Back to top