The first section of Chapter 6 in Principles of Gene Manipulation (pp 86-91) discusses the preparation of genomic DNA libraries. This is achieved by cutting genomic DNA randomly with restriction enzymes, followed by size fractionation, and insertion of the resulting fragments into appropriate vectors. Ideally, the library should cover the entire genome and the fragments should overlap.

However, genomic libraries require a relatively large amount of good-quality genomic DNA as starting material. It is difficult to prepare libraries from small amounts of material (e.g. single cells) and from material in which the DNA has degraded or has been fixed. The polymerase chain reaction (PCR) is ideal for such material because the technique is robust and can, at least in theory, amplify a single target molecule. Two PCR approaches were described in 1992 for whole genome amplification. The first strategy is known as primer extension preamplification (PEP) (Zhang et al. 1992). In this technique, genomic DNA is digested with restriction enzymes and the resulting fragments are ligated to linkers. PCR is then carried out using primers that anneal to the linkers. The second strategy is known as degenerate oligonucleotide primed PCR (DOP-PCR) (Telenius et al. 1992, Cheung & Nelson 1996) and employs a collection of random oligonucleotide primers to achieve non-specific amplification of genomic DNA. Size selection occurs in two stages. The upper size limit for amplification products is dictated by the PCR reaction itself, since only fragments up to about 3 kb are produced. The lower size limit is chosen by size selection prior to cloning.

Although both of the techniques are powerful, neither can actually amplify the entire genome. Rather, a sample of short genomic fragments is amplified, which can be useful for genotyping and similar applications but not for the creation of truly representative genomic libraries. A third method has been described more recently, based upon the concept of strand displacement amplification (SDA) using DNA polymerase from bacteriophage Φ 29, which undergoes rolling circle replication (Dean et al. 2001, reviewed by Hawkins et al. 2002). Like the PCR, SDA is a cyclical process that results in an exponential increase in the number of copies of the DNA target. However, unlike the PCR, SDA is isothermic (all reaction stages are carried out at the same temperature). Essentially what happens is that a primer anneals and is extended as in the PCR. However, instead of heating the reaction mixture to separate the nascent strand from the template, a restriction enzyme is used to cleave at the edge of the primer. The primer can then be extended once again, concurrently displacing the previously synthesized strand. This displaced strand can also act as a template in the next reaction cycle, and so on and so forth until the reaction is complete.

The use of random primers for SDA allows the rapid amplification of genomic DNA, as has been shown using human DNA as the template. Up to 30 μg of amplification product can be obtained from very limiting amounts of starting material, equivalent to less than five copies of the human genome. Further advantages of this technique include the greater accuracy and processivity of Φ 29 polymerase compared to PCR enzymes such as Taq and Pfu polymerases, resulting in genomic fragments of up to 10 kb in length with an error frequency of less than 1 in 10⁶. Comparisons of SDA-derived and traditional clone-based genomic libraries have confirmed that SDA does generate a representative set of fragments. Shotgun libraries were prepared from genomic DNA isolated in the normal manner from large cell cultures of the bacterium Xyella fastidiosa and from DNA produced by SDA from fewer than 1000 X. fastidiosa cells (without DNA isolation). In both cases, about 3000 sequence reads were obtained and aligned to the X. fastidiosa genome sequence. Taking into account the amount of sequence generated and the size of the X. fastidiosa genome, about 40% coverage was expected if the libraries were representative. Both libraries provided the expected coverage: 39% for the clone library and 34% for the SDA library (Hawkins et al. 2002 and references therein).

References:

Cheung V, Nelson S (1996) Whole genome amplification using a degenerate oligonucleotide primer allows hundreds of genotypes to be performed on less than one nanogram of genomic DNA. Proc Natl Acad Sci USA 93, 14676-9.

Dean F, Nelson J, Giesler T, Lasken R (2001) Rapid amplification of plasmid and phage DNA using Φ 29 DNA polymerase and multiply-primed rolling circle amplification. Genome Res 11, 1095-9.

Hawkins TL, Detter JC, Richardson PM (2002) Whole genome amplification - applications and advances. Curr Opin Biotechnol 13, 65-7.

Telenius H, Carter N, Bebb C, Nordenskjold M et al. (1992) Degenerate oligonucleotide-primed PCR: general amplification of target DNA by a single degenerate primer. Genomics 13, 718-25.

Zhang L, Cui X, Schmitt K, Hubert R et al. (1992) Whole genome amplification from a single cell: implications for genetic analysis. Proc Natl Acad Sci USA 89, 5847-51.