Asymmetric introgressive hybridization among louisiana iris species

Abstract

In this review, we discuss findings from studies carried out over the past 20+ years that document the occurrence of asymmetric introgressive hybridization in a plant clade. In particular, analyses of natural and experimental hybridization have demonstrated the consistent introgression of genes from Iris fulva into both Iris brevicaulis and Iris hexagona. Furthermore, our analyses have detected certain prezygotic and postzygotic barriers to reproduction that appear to contribute to the asymmetric introgression. Finally, our studies have determined that a portion of the genes transferred apparently affects adaptive traits.

Figure 1

The distribution of genetic markers among adult plants and seeds collected from an I. fulva x I. brevicaulis natural hybrid population. The genotypic score was based upon nuclear and chloroplast DNA markers. Individuals with scores of “0” or “9” indicated I. fulva or I. brevicaulis individuals, respectively. Those adult plants or seeds with scores from 1-8 were hybrids [20]

Figure 2

The observed and expected genotypic distributions at the L180 RAPD locus in I. fulva x I. brevicaulis F₂ progeny derived from crosses with either I. fulva (“F2f”) or I. brevicaulis (“F2b”) as the female parent [23].

Figure 3

The observed frequencies of introgressed alleles from either I. fulva (red lines) or I. brevicaulis (blue lines) into first generation backcross progeny formed from crosses between these two species. The X-axis indicates the genetic distances (in centimorgans) along each of the 21 linkage groups in the composite map. The Y-axis indicates the transmission ratio of either the I. fulva alleles or I. brevicaulis alleles introgressed into the backcrosses toward the alternate species. The expected frequency is 0.50 and is indicated by the dotted line. Data points above and below the solid lines indicate significant deviations from 0.50 (α = 0.05). Frequencies > 0.50 indicate an overrepresentation of either the I. fulva (red line) or I. brevicaulis (blue line) alleles in the genetic background of the alternate species. Frequencies < 0.50 indicate an underrepresentation of these same categories [24].

Figure 4

Schematic illustration of the distribution of 1) naturally occurring I. fulva plants (red ovals), 2) introduced I. hexagona (blue rectangle) and 3) I. fulva x I. hexagona F₁ plants (purple squares) [27,28].

Figure 5

Percentage of F₁ (0.03% and 0.74% in I. fulva and I. hexagona fruits, respectively) and first generation backcross seeds (B_f and B_h) formed on plants in an experimental population by natural pollinations [27,28]. The B_f and B_h hybrid seeds reflect the first generation of introgression into I. fulva and I. hexagona, respectively.

Figure 6

Percent F₁ hybrid seeds produced by various mixtures of I. fulva and I. hexagona pollen. The solid line illustrates the expected percentage of hybrid seeds assuming random fertilization. All the observed F₁ percentages were significantly less than expected (except for the 0% and 100% treatments, in which there were no mixtures of conspecific and heterospecific pollen). The blue and red rectangles indicate the percentage of heterospecific pollen necessary to increase significantly F₁ hybrid formation above the value of “0” in I. hexagona and I. fulva fruits, respectively. Note the much greater frequency of F₁s formed in I. hexagona fruits relative to I. fulva fruits [29].

Figure 7

Spatial distribution of Louisiana Iris genotypes in a natural population containing “I. brevicaulis-like” and “I. fulva-like” genotypes. Each circle reflects a single plant. The numbers indicate elevations, with the “0” line indicating the water level of the pond. Negative values reflect flooded areas, and positive values reflect areas above the waterline [19].