Exploitation of Solanum chilense and Solanum peruvianum in tomato breeding for resistance to Tomato yellow leaf curl disease

Abstract

Among viral diseases affecting cultivated tomato, Tomato yellow leaf curl disease (TYLCD) is one of the most devastating. This disease is caused by a complex of viruses of which Tomato yellow leaf curl virus (TYLCV) is regarded as the most important species. Current control strategies to fight viral diseases in tomato are mainly based on genetic resistance derived from wild relatives. In the present thesis, resistance derived from S. chilense and S. peruvianum has been exploited in breeding for resistance to TYLCD. In a previous study, TYLCV-resistant breeding lines derived from LA1932, LA1960 and LA1971 S. chilense accessions were developed. Therefore, the first objective of this thesis was to study the genetic control of the resistance derived from these accessions. With this aim, response to viral infection was assayed in segregating generations derived from the aforementioned resistant lines. The results obtained were compatible with a monogenic control of resistance. Resistance levels were higher in LA1960- and LA1971-derived F2 generations, as shown by slighter symptoms in the resistant plants and a higher number of asymptomatic plants compared with the results obtained in the LA1932-derived F2 generation. It is noteworthy that the level of resistance present in our materials is comparable to or even higher than the levels found in tomato lines homozygous for Ty-1. The response in plants heterozygous for the resistance gene was comparable to the response in homozygous plants for all three sources employed. This implies that the resistance genes derived from all three sources seem to be almost completely dominant. This effect was stronger for LA1971-derived resistance. The results were similar when comparing viral accumulation, as was expected, since a positive correlation was found in these families between viral accumulation and symptom scores. This has important implications in breeding, since the resistance will be used mostly for hybrid development. Our second objective was to map the loci associated with the major resistance genes identified. A total of 263 markers were screened, 94 of them being polymorphic between both species. Recombinant analysis allowed the resistance loci to be localized on chromosome 6, in a marker interval of 25 cM. This interval includes the Ty-1/Ty-3 region, where two S. chilense-derived TYLCD resistance loci were previously mapped. In order to test if the resistance genes identified in our populations were allelic to Ty-1 and Ty-3, further fine mapping was carried out. A total of 13 additional molecular markers distributed on chromosome 6 allowed 66 recombinants to be identified, and the resistance region to be shortened to a marker interval of approximately 950 kb, which overlaps with the Ty-1/Ty-3 region described previously by other authors. Therefore, the results obtained indicate that closely linked genes or alleles of the same gene govern TYLCV resistance in several S. chilense accessions. The third objective of the present thesis was to start the construction of a set of introgression lines (ILs) derived from Solanum peruvianum accession PI 126944 into the cultivated tomato genetic background. Once this collection of ILs is developed, it will represent a powerful tool for exploiting the resistance to different pathogens found in this particular accession in addition to other possible characters of interest. The starting plant material consisted of several segregating generations that were derived from two interspecific hybrids previously obtained by our group. Many crosses and embryo rescue were required to obtain subsequent generations due to the high sexual incompatibility that exists between tomato and PI 126944. Several mature fruits from the most advanced generations produced a few viable seeds, although embryo rescue was also employed to obtain progeny. As only a few plants were obtained by direct backcrossing, additional crosses were made in order to increase the number of descendants. A high degree of incompatibility was also found in crosses between sib plants. A total of 263 molecular markers were tested in some generations, 105 being polymorphic between tomato and PI 126944. Available generations were genotyped with these polymorphic markers in order to determine which alleles of S. peruvianum were already introgressed. On average, 79, 78 and 84 % of the S. peruvianum genome was represented in the pseudo-F2, pseudo-F4 and pseudo-F5 generations, respectively, for the markers analyzed. A reduction in the S. peruvianum genome was observed in more advanced generations, such as BC1 (56 %), pseudo-F2-BC1 (60 %) and pseudo-F3-BC1 (70 %). A greater reduction was observed in the pseudo-F3-BC2 generation (33 %). As a consequence of the reduction in the S. peruvianum genome, a loss of incompatibility was observed in some cases. The S. peruvianum genome was almost completely represented among the different plants of the most advanced generations. An evaluation for resistance to TYLCD and Tomato spotted wilt virus (TSWV) was carried out in some of the advanced generations, some of which were resistant to one or both viruses. In conclusion, we have conducted a successful and deeper exploitation of two wild species with proved resistance to TYLCD, S. chilense and S. peruvianum, identifying and fine mapping new genes of resistance.