Unravelling the genetic base of the meiotic recombination landscapes in two varieties of the button mushroom, Agaricus bisporus

The button mushroom, Agaricus bisporus var. bisporus, is one of the most cultivated mushrooms worldwide. Even though wild isolates of this variety have a broad genetic variation, the traditional and present-day hybrids only have a very narrow genetic base. The button mushroom has a typical meiotic recombination landscape (MRL) in which crossover (CO) events are predominantly restricted to the extreme ends of the chromosomes. This has been one of the main obstacles for mushroom breeders in improving or generating new mushroom hybrids due to a considerable linkage drag. A wild variety of A. bisporus, i.e., burnettii appeared to have CO spread more evenly across the genome. The existence of two extremely different MRLs in two compatible A. bisporus varieties offers an excellent opportunity to study the genetic basis for positioning CO in meiosis. The main objective of the research presented in this thesis initially was to examine meiosis of the var. burnettii in more detail and subsequently to identify genomic regions revealing the difference in MRL of the two A. bisporus varieties. The availability of genome sequences in the bisporus variety has produced many more informative markers such as SNP. We aimed to de novo sequence one of the haplotypes of a heterokaryotic strain of the burnettii variety using the PacBio sequencing technique and resequencing the other haplotype using Illumina HiSeq. In parallel to this, we used Genotyping by Sequencing (GBS) to construct the first linkage map of the burnettii variety, showing a more or less even distribution of COs across the genome. The constructed linkage map has also proved to be a useful tool for de novo assembly of the burnettii variety genome sequence. In addition, we performed comparative genome sequence studies between the burnettii variety and the previously sequenced genomes of two of the bisporus variety homokaryons, indicating high levels of collinearity between all three genomes. The only chromosomal rearrangement to be found was on chromosome 10, where an inversion of ~ 800 kb in the burnettii variety was detected compared to the var. bisporus genomes. As a starting point for unravelling the genetic basis underlying MRL in the A. bisporus, we performed quantitative trait loci (QTL) analysis using bisporus and burnettii varieties. An inter-varietal population was developed from a cross between a constituent nucleus of the bisporus and the burnettii variety. This population contains 178 haploid progenies which were genotyped by 210 SNP markers to construct a genetic linkage map, which proves to be a solid foundation for exploring the genetic control of MRL of A. bisporus. In addition, we performed a comparative genetic mapping study using the genetic maps of the bisporus variety Horst U1, the burnettii variety Bisp119/9 and the inter-varietal hybrid by selecting markers having similar positions in these three maps. In contrast to the bisporus variety where CO events are mainly restricted to chromosome ends, the burnettii variety shows a more or less equal distribution of CO events across the entire genome. The recombination landscape of the inter-varietal hybrid shows an intermediate pattern to that of both varieties. The MRL trait is expressed as a CO event in the offspring of each individual of the inter-varietal mapping population. For this reason, the individuals of the inter-varietal mapping population were intercrossed and outcrossed to generate three types of second generation hybrids. Two compatible tester homokaryons derived from the bisporus and burnettii varieties were used for outcrossing. Subsequently, the haploid progenies from each type of second generation hybrids were isolated to generate three types of segregating populations. The haploid progenies from segregating populations were genotyped with SNP markers covering the whole length of all the chromosomes. Recombination frequencies were determined at distal ends and elsewhere on the chromosomes and used to compare recombination frequencies between chromosomes within each population as well as between segregating populations across all chromosomes. A prerequisite for successful QTL mapping the MRL is to select segregating populations in which the segregation of MRL is clear. We observed that segregating populations outcrossed with the bisporus tester homokaryon were the most useful populations to generate haploid offspring in which COs are assessed for further QTL study of MRL at the time when this research was carried out. To map genomic regions involved in the different MRLs of A. bisporus, 71 homokaryotic offspring of the inter-varietal hybrid were outcrossed with an unrelated tester homokaryon of the bisporus variety. Subsequently, the haploid progenies were isolated from each hybrid and genotyped with SNP markers. Marker pairs were generated for the end regions of chromosomes to assess CO there or anywhere else on the chromosomes for each segregating population. QTL mapping analysis revealed two QTLs located on chromosome l and three others located on chromosomes IV, VI and VII. The QTLs identified span large parts of their respective chromosomes; therefore further strategies are needed for a more precise assessment and localisation of MRL.