SEARCH

SEARCH BY CITATION

Keywords:

  • drought tolerance;
  • hybridization;
  • introgression;
  • natural selection;
  • plant adaptation;
  • plant evolution

Interspecific hybridization has long been considered an important evolutionary mechanism in plants (Arnold, 2004). In particular, the transfer of adaptations among related taxa mediated by interspecific hybridization has received considerable attention from theoretical and empirical points of views. This process, often called ‘adaptive introgression’, involves the transfer of fitness-increasing alleles and their phenotypic effects from one taxon to another (Whitney et al., 2006; Fig. 1). Although many empirical examples of this process have been proposed (Rieseberg & Wendel, 1993; Table 1), and some of them convincingly demonstrated using appropriate empirical approaches (e.g. Martin et al., 2006; Whitney et al., 2006), it remains difficult to evaluate the overall importance of this process between related taxa. Moreover, even in documented cases, it is typically unknown whether multiple traits are involved in adaptive introgression, and which traits are more likely to introgress. Theoretical studies indicate that under heterogeneous environments, a single allele that is at an advantage in the alternative environment and genetic background is expected to introgress readily (Barton, 2001). However the transfer of multiple traits in introgressed populations, possibly with complex epistatic interactions, is not straightforward and depends heavily on conditions such as the rate and duration of hybridization and the detailed genetics of hybrid fitness (Barton, 2001). In this issue of New Phytologist, Whitney et al. (pp. 230–239) specifically address the issue of adaptive introgression of multiple traits by comparing phenotypes and fitness of natural and synthetic hybrids of two wild sunflower species, Helianthus annuus and Helianthus debilis. Together with the results obtained in a previous investigation (Whitney et al., 2006), they provide experimental evidence that introgression has modified several aspects of the recipient taxon’s phenotype in an adaptive manner, including traits associated with adaptation to both abiotic (drought tolerance) and biotic (herbivore resistance) features of the environment.

image

Figure 1.  Adaptive introgression under positive selection for drought tolerance. The recipient taxon has a wide distribution and comes into secondary contact with a drought-tolerant related taxon. Hybridization occurs between the two taxa, and introgressed populations of the recipient taxon establish themselves in a new xeric environment. These populations have evolved adaptations to their environment that correspond to drought tolerance traits (e.g. an increase in the extent of branching) introgressed from the donor parental taxon to the recipient taxon.

Download figure to PowerPoint

Table 1.   Examples of types of traits for which events of adaptive introgression have been demonstrated or suggested; the traits fall into three main categories, depending on the nature of the responsible selective forces
Nature of selectionRelevance of traitsExample taxaReferences
Adaptation to the abiotic environmentShade toleranceIris fulva[RIGHTWARDS ARROW]Iris hexagonaReferences given in Arnold (2004)
Drought toleranceHelianthus debilis[RIGHTWARDS ARROW]Helianthus annuusWhitney et al. (2010, this issue)
Flood toleranceIris fulva[RIGHTWARDS ARROW]Iris brevicaulisMartin et al. (2006)
Adaptation to the biotic environmentHerbivore resistanceHelianthus debilis[RIGHTWARDS ARROW]Helianthus annuusWhitney et al. (2006)
Invasiveness associated with vegetative growthReynoutria sachalinensis[LEFT RIGHT ARROW]Reynoutria japonicaPysek et al. (2003)
Mating system dynamicsOutcrossing device (ray floret)Senecio squalidus[RIGHTWARDS ARROW]Senecio vulgarisKim et al. (2008)
Self-incompatibility allelesArabidopsis lyrata[LEFT RIGHT ARROW]Arabidopsis halleriCastric et al. (2008)

‘… adaptive introgression, coupled with some form of balancing selection, might constitute a recurrent evolutionary scenario.’

Methodological approaches to demonstrate adaptive introgression

  1. Top of page
  2. Methodological approaches to demonstrate adaptive introgression
  3. Evolutionary implications of adaptive introgression
  4. References

The ultimate demonstration of adaptive trait introgression would be based on evidence that a given haplotype originating from the donor parental species, and with a well-defined phenotypic effect, is subject to positive selection in a hybrid lineage (Martin et al., 2006; Whitney et al., 2006). A recent study by Kim et al. (2008) approached this standard, reporting the transfer of a cluster of regulatory genes (the RAY locus) from an introduced species Senecio squalidus to the UK-native species Senecio vulgaris. This study provided molecular evidence for introgression of this gene cluster into S. vulgaris, and showed that this event was associated with major changes in inflorescence morphology (introducing peripheral flowers with large conspicuous petals), possibly increasing outcrossing rates in this self-compatible species. Precise identification of the molecular basis of the adaptive trait allowed alternative processes, such as incomplete lineage sorting (ancestral origin of the trait maintained in both species since speciation) or parallel evolution (convergent evolution of the same trait), to be ruled out. However, in this case evidence for positive selection on the introgressed haplotype in S. vulgaris was missing.

When the molecular basis of the putatively introgressed trait is not known, robust validation of the adaptive introgression event requires a combination of several lines of evidence (Whitney et al., 2006): evidence for genetic introgression from the donor into the recipient taxon; evidence that introgressed lineages are maintained in the recipient taxon; evidence that phenotypes of the recipient taxon at fitness-related traits have shifted towards those of the donor; and evidence that natural selection has favored recipient genotypes with phenotypes shifted towards those of the donor. In the case of H. annuus and H. debilis, the transfer of genetic material from the donor into the recipient taxon was confirmed using microsatellite markers (Scascitelli et al., 2010). There is also clear evidence that the introgressed taxon has a long history of occurrence in southern Texas, USA. Now, Whitney et al. provide experimental evidence that many phenotypic traits (ecophysiological, phenological and plant architectural traits) potentially related to drought tolerance differ significantly between introgressed and nonintrogressed populations of the recipient taxon H. annuus. In all cases, values of these traits in the introgressed populations were shifted towards the phenotype of the donor H. debilis. These comparisons were replicated in two ecologically distinct sites. Finally, two synthetic BC1 (first generation of backcross with the recipient taxon) hybrid populations were used to specifically test for an association between these traits and individual fitness (as estimated by viable seed production) using the approach of phenotypic selection analysis (Lande & Arnold, 1983). The results highlight that selection in a dry habitat has favored phenotypes in the resynthesized hybrids that are shifted towards those of H. debilis. Altogether these results carefully demonstrate that adaptive introgression has resulted in a transfer of traits, allowing rapid growth and reproduction before summer heat and drought, from H. debilis to the genomic background of H. annuus. More importantly, they show that multiple traits related to adaptation to abiotic features of the environment can be transferred together.

Evolutionary implications of adaptive introgression

  1. Top of page
  2. Methodological approaches to demonstrate adaptive introgression
  3. Evolutionary implications of adaptive introgression
  4. References

From the many instances of adaptive introgression suggested in the empirical literature, actual demonstrations using the criteria already described are limited to a few reference cases (e.g. Louisiana Iris species (Martin et al., 2006) and sunflower (Whitney et al., 2006, 2010)). It is now essential to investigate this process in other taxa, and for a wide diversity of fitness-related traits, before concluding about the overall evolutionary importance of adaptive introgression. Future successful investigations would combine population genomic approaches and functional analyses, in order to validate the introgression process and to identify the molecular basis of adaptation, with phenotypic selection experiments aimed at demonstrating that the introgression process has been driven by natural selection.

Interestingly, several cases of adaptive introgression have been suggested recently for traits that are subsequently maintained as polymorphisms within the recipient taxon. One case corresponds to the transfer of the radiate inflorescence form from S. squalidus to S. vulgaris (see earlier). This introgressed trait seems to be maintained as a stable polymorphism within the recipient taxon and contributes potentially to an increase in population levels of outcrossing (Kim et al., 2008). A second case corresponds to the transfer of individual alleles at the multiallelic self-incompatibility locus between Arabidopsis lyrata and Arabidopsis halleri (Castric et al., 2008; Fig. 2). High allelic diversity at this locus is maintained within species through negative frequency-dependent selection (advantage of the rare), but hybridization has caused transfers of alleles between species driven by natural selection. Outside the plant literature, a third case has been described corresponding to a transfer of a mutation from domestic dogs to North American gray wolf populations, causing skin color change (melanistic K locus mutation; Anderson et al., 2009). These examples suggest that adaptive introgression, coupled with some form of balancing selection, might constitute a recurrent evolutionary scenario. Interestingly, in each case, the introgressed allele did not seem to have evolved in the donor parental taxon but was inherited from an ancestor, yet it was absent or lost in the lineage leading to the recipient species. Hence, it has been argued that adaptive introgression is an important mechanism for the long-term maintenance of many alleles in multiallelic systems subject to balancing selection (e.g. plant self-incompatibility; Schierup & Vekemans, 2008), or for the recovery of ancestral characters that may have been lost in some lineages (e.g. ray florets in Senecio; Kim et al., 2008).

image

Figure 2.  Adaptive introgression under negative frequency-dependent selection of an allele at the self-incompatibility locus. A pollen grain carrying a given allele at the self-incompatibility locus can only successfully fertilize individuals carrying a different allele (with a different color). Through hybridization, an allele specific to the donor parental taxon (light gray) is introduced (as a pollen grain) in the recipient taxon (double arrow) where it will benefit from the advantage of the rare as it can fertilize potentially all extant individuals (adapted from Schierup & Vekemans, 2008).

Download figure to PowerPoint

Finally, adaptive introgression may also have evolutionary implications as a process leading to the transfer of traits from domesticated gene pools to wild gene pools. A key concern is constituted by the risk for increased weediness associated with the potential transfer of herbicide- or insect-resistance alleles from genetically modified crops to wild populations of related species (Stewart et al., 2003). However, the North American gray wolf case also suggests that, for an endangered species, ‘hybridization with another species or a domesticated relative can provide genetic variation that has important adaptive consequences’ (Hedrick, 2009).

References

  1. Top of page
  2. Methodological approaches to demonstrate adaptive introgression
  3. Evolutionary implications of adaptive introgression
  4. References