Figure 2.5 Example of results from a differential screening approach applied to clones from a subtracted library (sexual versus asexual strains from the freshwater planarian G. tigrina). Clones arrayed on nylon membranes were selected randomly from the subtracted library (top panels) to study upregulation, and the reverse-subtracted library (bottom panels) to study downregulation. The dot-blots were then hybridized to labelled probes prepared from the subtracted (left-hand panels) and reverse-subtracted libraries (right-hand panels). After Rebrikov et al. (2004) by permission of Humana Press.
cube of the expression ratio. This non-linear effect implies that genes with highly differential expression (a high expression ratio) will be much more strongly enriched than genes with a lower expression ratio. For example, if genes with an expression ratio of 2 are enriched 10 times, genes with an expression ratio of 5 are enriched by a factor of 156. Theoretical and empirical arguments have demonstrated that only genes differing in expression by a factor of 5 or more will be effectively picked up in current SSH protocols.
Application of SSH in an ecological context is growing rapidly, especially because it is a good preparatory method for generating enriched cDNA clone libraries for spotting microarrays (see below). To illustrate the use of SSH in ecology, a study by Pearson et al. (2001) is exemplary. These authors screened for genes differentially expressed in the brown macroalga, Fucus vesiculosus (of the family Phaeophyceae), subjected to drought stress. Many genes were found to have differential expression that could not be identified by homo-logy to known sequences, but those that could included partial sequences for ribulose-1,5-bisphosphate carboxylase/oxygenase, chloroplast-coupling factor ATPase, and a photosystem I P700 chlorophyll a-binding protein. This study illustrates the great flexibility of genes encoding components of the photosynthetic apparatus, not only in response to light, but also to the hydration status of the tissues.
Various DNA-fingerprinting methods, which are very popular in molecular ecology to elucidate population structure, geographic variation, or paternity (Beebee and Rowe 2004), can often form the starting point of functional analysis. DNA fingerprinting in general can be done in two ways. One approach is to use identified, often non-coding, polymorphic sequences in the genome, such as microsatellites (loci with a variable number of short tandem repeats, e.g. (GA)n, where n varies from, let's say, 5 to 9). Since these are single-locus codominant markers (the heterozygotes can be distinguished from either homozygote), they are especially suitable for population analysis (Jarne and Lagoda 1996, Sunnucks 2000). Another approach is multilocus DNA fingerprinting, where genotypes are recognized from many markers at the same time, often of unknown sequence. One such multilocus analysis, popular at the beginning of the 1990s when the molecular approach in ecology began its advance, goes under the name of randomly amplified polymorphic DNA (RAPD). Williams et al. (1990) introduced this technique, which is based on a PCR with 10- to 15-mer primers of arbitrary sequence. Due to variation between individuals in the position or sequence of primer-annealing sites, each individual generates a different series of bands when PCR products are separated on an agarose gel. A large number of primers (sometimes several hundreds) is used to probe the genome, hence the designation 'random'.
When RAPD markers are found to be associated with certain environmental conditions, important ecological traits, or phenotypes of interest, they are cloned and sequenced, and new primers are developed based on the sequence, allowing a more robust PCR. The DNA segment is then designated with the awkward term sequence-characterized amplified region, SCAR, or the more general term sequence-tagged site, STS. The use of SCARs has become very popular in plant breeding and crop science, because it allows rapid screening of many individuals for certain traits of interest and it aids marker-assisted selection programmes. In such breeding programmes, SCARs are designed to link with resistance/susceptibility genes or other genes that determine the commercial value of the plant or animal. For example, Haymes et al. (1997) developed a SCAR for resistance of strawberry (Fragaria x ananassa) to the fungus Phytophthora fragariae (Oomycota), which causes a form of root rot. With a combination of primers, a reliable identification could be made for a resistance allele of the Rpf (regulation of pathogenicity factors) gene, which encodes a small excreted protein with a signalling function.
Because the RAPD procedure produces only a limited number of bands from each primer and the banding pattern is sensitive to the amount of template DNA and PCR conditions (e.g. Mg2 + concentration), more reliable fingerprinting techniques have been developed, one of the most popular being amplified fragment length polymorphism (AFLP; Vos et al. 1995). In this technique specific adapter sequences are ligated to DNA digests obtained with two restriction enzymes before a PCR is done. One of the restriction enzymes is a frequent cutter—it binds to a short, common sequence of nucleotides—that will ensure that sufficient fragments are obtained with a size range that allows easy separation by electrophoresis; the other is a rare cutter—it binds to a longer, less-common sequence—used to limit the number of fragments amplified by the PCR (only the fragments with a frequent cut on one side and a rare cut on the other side are amplified). The PCR uses primers targeted to the adapter sequences, with one, two, or three bases extending in the amplicon, to select a subpopulation of the fragments. The reaction products are resolved by electrophoresis on a polyacrylamide gel (Fig. 2.6). AFLP can also be applied to cDNAs, in which case it can be used as a differential screening method (see above) when the fingerprints from two pools of cDNA are compared for differential bands (cDNA-AFLP).
In complex genomes the number of bands obtained can be very large, sometimes leading to difficulty in interpretation when AFLPs are used for resolving population structure. The number may be decreased by extending the selective bases. For example, using an overhang of four rather than three selective bases, as in Fig. 2.6, reduces the expected number of bands by a factor of 16. Another strategy is the use of three rather than two endonucleases, while retaining only two adapters. Depending on the recognition sequence of the third enzyme, this leads to a reduction of the number of amplified fragments by a factor of
Was this article helpful?