Analysis of 41 plant genomes supports a wave of successful genome duplications in association with the Cretaceous-Paleogene boundary

Abstract

Ancient whole-genome duplications (WGDs), also referred to as paleopolyploidizations, have been reported in most evolutionary lineages. Their attributed role remains a major topic of discussion, ranging from an evolutionary dead end to a road toward evolutionary success, with evidence supporting both fates. Previously, based on dating WGDs in a limited number of plant species, we found a clustering of angiosperm paleopolyploidizations around the Cretaceous-Paleogene (K-Pg) extinction event about 66 million years ago. Here we revisit this finding, which has proven controversial, by combining genome sequence information for many more plant lineages and using more sophisticated analyses. We include 38 full genome sequences and three transcriptome assemblies in a Bayesian evolutionary analysis framework that incorporates uncorrelated relaxed clock methods and fossil uncertainty. In accordance with earlier findings, we demonstrate a strongly nonrandom pattern of genome duplications over time with many WGDs clustering around the K-Pg boundary. We interpret these results in the context of recent studies on invasive polyploid plant species, and suggest that polyploid establishment is promoted during times of environmental stress. We argue that considering the evolutionary potential of polyploids in light of the environmental and ecological conditions present around the time of polyploidization could mitigate the stark contrast in the proposed evolutionary fates of polyploids.

Figure 1.

K_S age distributions for (A) M. truncatula, (B) C. arietinum, (C) L. japonicus, (D) C. cajan, (E) A. thaliana, (F) S. lycopersicum, (G) O. sativa, and (H) M. acuminata. The gray and gold bars represent the distribution of the paranome and duplicated anchors identified with i-ADHoRe, respectively. Anchors and peak-based duplicates used as homeologs for absolute dating were extracted from between the WGD peak boundaries (see Table 1). The gray box surrounding A–D indicates that these four species represent the same Faboideae-specific WGD.

Figure 2.

Absolute age distributions of the dated anchors (AP, left) and peak-based duplicates (PB, right) for (A) M. truncatula, (B) C. arietinum, (C) L. japonicus, (D) C. cajan, (E) A. thaliana alpha duplication, (F) S. lycopersicum, (G) O. sativa, and (H) M. acuminata. (Nonvertical black solid line) Kernel density estimate of the dated homeologs; (vertical black solid line) its peak used as the WGD age estimate. (Gray solid lines) Density estimates for the 1000 bootstrap replicates; (vertical black dashed lines) corresponding 90% confidence intervals on the WGD age estimate. The original raw distribution of dated homeologs is also indicated on the individual plots by open dots. See Table 1 for sample sizes and exact confidence interval boundaries. The gray box surrounding A–D indicates that these four species represent the same Faboideae-specific WGD.

Figure 3.

Phylogenetic tree of the green plants incorporating all species used in this study, with the exception of N. nucifera, which as a public annotation was not yet available upon completion. In total, sequence information from 38 full genome sequences and three transcriptome assemblies was used (see Supplemental Table S1). Bars indicate all known WGDs. Black bars indicate WGD age estimates from the literature and are not to scale (see Supplemental Information for justification and corresponding references). (Green bars) Estimates for WGDs dated in this study, with right and left boundaries corresponding to the youngest and oldest 90% confidence interval boundary found in the complete set of species-specific WGD age estimates that descend from each independent WGD (see Table 1). Some WGDs in woody species such as G. raimondii (Malvales), P. trichocarpa, and M. esculenta (Malpighiales), and the WGD shared by both M. domestica and P. bretschneideri (Rosales), are most likely underestimated through strong rate deceleration that is not fully corrected for (see Results and Discussion; Supplemental Information). The fading brown bars for the WGD in P. trichocarpa, and the WGD shared by M. domestica and P. bretschneideri, indicate corrected WGD age suggestions based on fossil evidence and/or other dating studies (see Results and Discussion). The green bar for M. acuminata most likely represents two separate WGDs in close succession (D’Hont et al. 2012). A possible WGD at the base of the monocots is not indicated because its exact phylogenetic placement remains unclear (Paterson et al. 2004). Branch lengths are truncated after 150 mya to improve clarity.