High-density detection of restriction-site-associated DNA markers for rapid mapping of mutated loci in Neurospora

Abstract

The wealth of sequence information available for Neurospora crassa and other fungi has greatly facilitated evolutionary and molecular analyses of this group. Although "reverse" genetics, in which genes are first identified by their sequence rather than by their mutant phenotypes, serves as a valuable new approach for elucidating biological processes, classical "forward" genetic analysis is still extremely useful. Unfortunately, mapping mutations and identifying the corresponding genes has typically been slow and laborious. To facilitate forward genetics in Neurospora, we have adapted microarray-based restriction-site-associated DNA (RAD) mapping for use with N. crassa oligonucleotide microarrays. This technique was used to simultaneously detect an unprecedented number of genomewide restriction site polymorphisms from two N. crassa strains: Mauriceville and Oak Ridge. Furthermore, RAD mapping was used to quickly map a previously unknown gene, defective in methylation-7 (dim-7).

Figure 1.—

Schematic of the RAD mapping procedure. Genomic DNA of two polymorphic strains is digested with a restriction enzyme. In this study, we used the dim-7 (N3312) strain, which is predominantly an Oak Ridge-type strain, and the polymorphic Mauriceville (N32) wild-type strain. Biotinylated linkers are then ligated to the restriction sites, the DNA is sheared, and the RAD tags are purified using streptavidin beads. The purified tags from each strain are differentially labeled with Cy3 or Cy5 and then hybridized to a microarray. Shared restriction sites generate yellow spots, whereas strain-specific restriction sites generate red or green spots.

Figure 2.—

RAD tags are reproducibly detected. The log₂ (N3312 dim-7/N32 Mauriceville) values from the first parental experiment are plotted on the y-axis and the log₂ (N32 Mauriceville/N3312 dim-7) values from the second experiment are plotted on the x-axis. The negative slope is a result of reciprocal ratios produced in the dye swap experiment. The Pearson's correlation coefficient (r = −0.87) was calculated for all spots that showed a log₂ value ≤ −1 or ≥1.

Figure 3.—

Distribution of EcoRI–RAD markers across Neurospora chromosomes. Spots that showed a twofold or more enrichment in the dim-7 (Oak Ridge) or Mauriceville strains in two independent experiments were scored as informative RAD markers. The vertically arranged rectangles represent the seven N. crassa LGs. For each LG, dim-7-specific RAD markers are depicted as solid triangles on the right and Mauriceville-specific RAD markers are depicted as solid triangles on the left. Horizontal lines along the LG separate consecutive contigs. To generate a representation of EcoRI–RAD marker distribution, sequence files were assembled for each LG. Contig order was based on the physical map data available at the Neurospora gene list e-Compendium (http://www.bioinf.leeds.ac.uk/∼gen6ar/newgenelist/genes/physical_maps.htm). Contigs that cannot be given an orientation (e.g., a “+” or “−” strand) are shaded. Contigs that cannot be assigned to a specific LG are labeled “unmapped” at the bottom right. Contigs that can be placed on an LG but not ordered within the LG are shown beneath LGs V, VI, and VII. The position of each RAD marker was determined by aligning the microarray probe sequence with the assembled sequence files using BLAST (Altschul et al. 1990). A Perl script was written to parse the BLAST data and draw an image for each LG.

Figure 4.—

RAD mapping by bulk segregant analysis places dim-7 on LG V . RAD markers that showed a twofold or more enrichment in either the wild-type or the mutant pools are shown as black bars. The height of each bar indicates the relative enrichment of the RAD marker (bulk segregant experiment/parental). The RAD markers that did not show an enrichment of twofold or more in either the wild-type pool or the mutant pool are shown as gray triangles. The asterisk indicates the location of the NCU04209.1 array element. The 0.4-Mb bar indicates the location of dim-7 confirmed by traditional RFLP mapping. The numbers beneath the graph indicate contig numbers corresponding to the Neurospora Genome Sequence Assembly 7 (http://www.broad.mit.edu/annotation/genome/neurospora/Home.html). To generate the representation of the bulk segregant data, a sequence file for LG V was assembled. Contig order was based on the physical map data available at the Neurospora gene list e-Compendium (http://www.bioinf.leeds.ac.uk/∼gen6ar/newgenelist/genes/physical_maps.htm). Contigs containing rDNA repeats were not included in the figure. Contigs 46, 88, and 91 cannot be assigned an orientation (± strand). The position of each RAD marker was determined by aligning the microarray probe sequence with the LG V sequence files using BLAST (Altschul et al. 1990). A Perl script was written to parse the BLAST data and draw the image. (A) The data from individual replicate experiments. (B) The average value from replicates 1 and 2. Only probes that showed a twofold or more enrichment in both replicate experiments are plotted as black bars in B.