Chromosomal deletions (also called genetic deficiencies) have been used to help narrow down the location of genes in organisms ranging from C. elegans to human beings 1234. In the invertebrate model system, Drosophila melanogaster, deficiencies have been particularly useful for gene mapping, since the existence of polytene chromosomes and cytogenetics in this organism allows the actual visualization of the breakpoints of the deficiencies. In other organisms, however, mapping the breakpoints of deficiencies is not so straightforward. In C. elegans, for example, deficiencies are defined as deletion mutations that fail to complement multiple loci with the underlying assumption that the deletion would also remove the regions between these loci. Although this is generally true, deficiencies exist that are complex deletions, eliminating two distant loci, but retaining the intervening sequence (Figure 1). An example of such a complex deficiency is shown in Table 1. Genetic complementation tests 5 show that although the deficiency sDf62 deletes a region from map position 3.99 to 5.78 of chromosome IV, it fails to delete some genes that lie within this map interval (e.g. unc-43 and lin-3). In fact, our searches on WormBase 5 reveal that a surprising number of deficiencies in C. elegans are of a complex nature. When we examined a random sampling of 20 deficiencies from WormBase 5 we found that 9 of these were complex deficiencies. Although 20 is not, by any means, a large sample, it indicates that a sizeable number of deficiencies in C. elegans might be complex. This, coupled with the relative scarcity of usable genetic markers can confound the interpretation of deficiency data.

In order to address the problem of deletion breakpoint mapping in C. elegans, previous work 6 has attempted to use PCR to determine the breakpoints of such deficiencies (Figure 2). Large deletions in the homozygous state are usually lethal to the developing C. elegans embryo. Thus, from a plate of worms heterozygous for the deletion, one can pick individual dead eggs that have been laid, assuming that these must be homozygous for the deletion. One can then use PCR to test whether a particular region of interest is present in such samples, thereby determining the extent of the deletion. This method, which relies on a negative result (i.e., lack of a PCR product), has several drawbacks: (a) The amount of starting material is restricted to a single, or few dead eggs. (b) Chitinase treatment is required to remove the egg shell. (c) False positives can result if the PCR merely fails to work properly on the limited sample material. (d) False negatives can result if an embryo dies for reasons other than that it was homozygous for the deletion.

Another method that could be used to achieve the same result is Southern Blotting 7. However, this method is tedious, involving the preparation of high molecular weight genomic DNA, running gels and probe hybridization, and requires a larger starting amount of DNA.

In human research, taking advantage of the growing number of SNPs identified in the human genome, approaches are being developed to use SNPs to identify regions of loss of heterozygosity (LOH), and/or deletions in chromosomes 8. The availability of a strain of C. elegans, CB4856, which is predicted to harbor about 1 SNP per kilobase of genomic sequence 910 when compared to the wild-type strain (N2), led us to believe that we could adopt similar methods for use in C. elegans, thereby overcoming the limitations of using deficiencies in this model system. In fact, the C. elegans SNP database already contains about 6,000 confirmed and predicted SNPs between the N2 and CB4856 strains 11, which corresponds to a SNP density of about 1 per 15 Kb of sequence. As the use of SNPs to map genes increases 1213, we expect that many more SNPs will be discovered, making their use even more attractive. We report here the use of SNP mapping, in conjunction with traditional genetic deletion mapping to obtain a better map of the breakpoints of a complex deficiency, thereby greatly assisting the process of gene mapping and characterization.

Our use of a combination of genetic complementation data along with SNP mapping of deficiency breakpoints has allowed us to rapidly narrow down the region of interest for cloning spe-19. Deficiency mapping in a number of organisms, including C. elegans, has been beset by the lack of a convenient method to correlate genetic complementation data with the physical map, confounding our ability to accurately define the breakpoints of deficiencies. Previous methods used to overcome this obstacle depended on a negative result 67, which is often untrustworthy. Using SNPs, not only is the correlation between the genetic and physical maps easily achieved, but we can also, potentially, generate higher resolution maps of the breakpoints of both simple, as well as complex deficiencies. Additionally, the amplification of the identical locus from the CB4856 chromosome serves as a good internal control, eliminating false positives and false negatives and allowing us to interpret our data with a very high degree of confidence. The fact that we can use heterozygous animals, without the need to generate homozygous, dead embryos overcomes the limitations of previous methods which relied exclusively on negative results, such as using PCR to detect the presence or absence of a sequence. We chose to use bulk worm lysates from a heterozygous line as our source of DNA. However, the technique would work just as well if single worms (after confirming their genotype) were used as the source of DNA. We believe that using lines has the following advantages:

1. The line can be frozen away and then rethawed at a later date if further analysis is required.

2. Multiple SNPs can be tested using the same lysate, as opposed to requiring multiple, single worm lysates.

3. Lysates can be prepared in bulk, in replicates, for independent and robust confirmation of all results obtained by this method.

4. No special care needs to be taken for preparation of DNA for SNP detection. DNA is made from the lines using a simple, scaled up version of the method used to make worm lysates for single worm PCR.

We initially chose to use SNPs that caused an abrogation of a restriction enzyme recognition site in the CB4856 sequence, thereby eliminating any artefacts due to incomplete restriction enzyme digestion. However, by running a control reaction to ensure that enzyme digestion has progressed to completion, one could also use SNPs that cause the loss of a restriction enzyme recognition site in the N2 sequence. Additionally, this method allows the use of SNPs that are only different in sequence, causing no changes in restriction enzyme recognition sites. A simple sequencing reaction can be done to check whether the Deficiency/CB4856 worms had only CB4856 sequence (which would be the case if the deficiency deletes the region of the SNP), or whether both N2 and CB4856 sequence was present (indicating that the SNP region is not lost in the deletion being tested). In addition to the CB4856 strain which we have used, other wild isolates of C. elegans which harbour SNPs 9 can also be used, expanding the possibilities of this technique.

While we have shown this method to work in the model system C. elegans, a similar approach can be applied in any organism where deficiencies/deletions exist and where SNP mapping is possible.

PK conceived of using SNPs to analyse deletion breakpoints, carried out the SNP analysis and drafted the manuscript. All of the genetic complementation mapping was carried out by BG. MH aided with the molecular biology. AS participated in designing the experiments and in their coordination and helped to draft the manuscript. All work was conducted in the laboratory of AS.