Evidence of cryptic introgression in tomato (Solanum lycopersicum L.) based on wild tomato species alleles

Abstract

Background: Many highly beneficial traits (e.g. disease or abiotic stress resistance) have been transferred into crops through crosses with their wild relatives. The 13 recognized species of tomato (Solanum section Lycopersicon) are closely related to each other and wild species genes have been extensively used for improvement of the crop, Solanum lycopersicum L. In addition, the lack of geographical barriers has permitted natural hybridization between S. lycopersicum and its closest wild relative Solanum pimpinellifolium in Ecuador, Peru and northern Chile. In order to better understand patterns of S. lycopersicum diversity, we sequenced 47 markers ranging in length from 130 to 1200 bp (total of 24 kb) in genotypes of S. lycopersicum and wild tomato species S. pimpinellifolium, Solanum arcanum, Solanum peruvianum, Solanum pennellii and Solanum habrochaites. Between six and twelve genotypes were comparatively analyzed per marker. Several of the markers had previously been hypothesized as carrying wild species alleles within S. lycopersicum, i.e., cryptic introgressions.

Results: Each marker was mapped with high confidence (e<1 x 10-30) to a single genomic location using BLASTN against tomato whole genome shotgun chromosomes (SL2.40) database. Neighbor-joining trees showed high mean bootstrap support (86.8 ± 2.34%) for distinguishing red-fruited from green-fruited taxa for 38 of the markers. Hybridization and parsimony splits networks, genomic map positions of markers relative to documented introgressions, and historical origins of accessions were used to interpret evolutionary patterns at nine markers with putatively introgressed alleles.

Conclusion: Of the 47 genetic markers surveyed in this study, four were involved in linkage drag on chromosome 9 during introgression breeding, while alleles at five markers apparently originated from natural hybridization with S. pimpinellifolium and were associated with primitive genotypes of S. lycopersicum. The positive identification of introgressed genes within crop species such as S. lycopersicum will help inform conservation and utilization of crop germplasm diversity, for example, facilitating the purging of undesirable linkage drag or the exploitation of novel, favorable alleles.

Figure 1

Chromosomal map locations of 47 markers sequenced in this study. Nine markers with dashed outlines showed cryptic introgressions. Also shown on the map (color) are documented introgressions used in tomato breeding that were mentioned in this report, S. habrochaites: blue, S, lycopersicum Peru Wild: red, S. peruvianum: purple, S. pimpinellifolium: orange, S. pennellii: green.

Figure 2

Examples of splits networks at markers with cryptic introgressions inS. lycopersicum. Dashed boxes indicate accessions with introgressed alleles (a-d: hybridization networks using marker U146437 as a control, e: parsimony splits network); a) TA496 showed reticulation with S. peruvianum LA1537 and other green-fruited species, b) TA496 showed reticulation with red-fruited and green-fruited species, c) TA496 nested within green-fruited species, d) Merville des Marchés PI 109834 was closely related to S. pimpinellifolium, which showed reticulation to Tomate PI 99782, e) parsimony splits network used only conservative, global signal within the marker to illustrate the hybrid origin of Merville des Marchés PI 109834.

Figure 3

Extreme divergence of marker C2_At1g73180.a) Clustering at marker C2_At1g73180 showed the distinctiveness of PI 129026 and PI 196297 with 99% bootstrap support, b) the splits network showed a combination of introgression of PI 129026 and PI 196297 (among others, see Table 2) with S. pimpinellifolium and retention of ancient polymorphisms that reticulated down to the root; the latter supported genetic hitchhiking and rejection of selective neutrality.