Ancestral grass karyotype reconstruction unravels new mechanisms of genome shuffling as a source of plant evolution

Abstract

The comparison of the chromosome numbers of today's species with common reconstructed paleo-ancestors has led to intense speculation of how chromosomes have been rearranged over time in mammals. However, similar studies in plants with respect to genome evolution as well as molecular mechanisms leading to mosaic synteny blocks have been lacking due to relevant examples of evolutionary zooms from genomic sequences. Such studies require genomes of species that belong to the same family but are diverged to fall into different subfamilies. Our most important crops belong to the family of the grasses, where a number of genomes have now been sequenced. Based on detailed paleogenomics, using inference from n = 5-12 grass ancestral karyotypes (AGKs) in terms of gene content and order, we delineated sequence intervals comprising a complete set of junction break points of orthologous regions from rice, maize, sorghum, and Brachypodium genomes, representing three different subfamilies and different polyploidization events. By focusing on these sequence intervals, we could show that the chromosome number variation/reduction from the n = 12 common paleo-ancestor was driven by nonrandom centric double-strand break repair events. It appeared that the centromeric/telomeric illegitimate recombination between nonhomologous chromosomes led to nested chromosome fusions (NCFs) and synteny break points (SBPs). When intervals comprising NCFs were compared in their structure, we concluded that SBPs (1) were meiotic recombination hotspots, (2) corresponded to high sequence turnover loci through repeat invasion, and (3) might be considered as hotspots of evolutionary novelty that could act as a reservoir for producing adaptive phenotypes.

Figure 1.

Ancestral grass karyotype reconstruction. The monocot (rice, Brachypodium, sorghum, maize) chromosomes are represented with color codes to illustrate the evolution of segments from a common ancestor with five protochromosomes (named according to the rice nomenclature A5, A4, A7, A8, A11). The current structure of the four genomes is represented at the bottom of the figure, with the seven ancestral duplications highlighted with gray boxes. Large segmental inversions are indicated with red arrows in Brachypodium, sorghum, and maize genomes according to the synteny, with rice used as reference genome. The ancestor intermediate (n = 12 for Pooideae, A1 to A12; and n = 10 for Panicoideae, A1 to A10) is illustrated by blocks of reordered genes, with the chromosome and duplication five-color code described above. The AGK (top), structured in seven blocks of five protochromosomes, went through a WGD and two chromosome fusions and fissions (A3 = A7 + A10; A2 = A4 + A6) to reach the n = 12 ancestor intermediate. Polyploidization events are indicated as WGD in the figure. The number of reordered genes, gene blocks, NCF, and CI events are indicated on the depicted modern and ancestral genomes.

Figure 2.

Synteny break points characterization. (A) The four Brachypodium (b1-2-3-4) and two sorghum (s1-2) chromosomes harboring the nine NCFs (black dots linked with dotted lines) identified in these genomes are illustrated according to their ancestral chromosomal origin (i.e., the A1 to A12 color code provided as the AGK painting scale): NCF#1-2 = A6-7-3, NCF#3 = A1-5, NCF#4-5 = A10-8-2, NCF#6-7 = A11-9-12, NCF#8 = A3-10, and NCF#9 = A7-9. For the six chromosomes, a heat map is provided for the telomeric repeat (blue, 0; yellow, <40%; red, >40%), centromeric repeat (blue, 0; yellow, <40%; red, >40%), LTR TE (blue, <80%; yellow, >80%; red, ∼100%), CDS (blue, <40; yellow, 40–50; red, >50), and CNV (blue, <3; yellow, 3–5; red, >5) distribution. The radars for Brachypodium and sorghum genomes represent the percentage (from 0 to 100%) for the centromeric, telomeric, SBPs, and internal and external NCF sequence regions (corresponding to the five radar peaks) harboring centromeric (blue), telomeric (purple), and TE (yellow) repeats as well as CDS (light blue). (B) Detailed representation of the NCF#19. The microsynteny is illustrated for the sorghum chromosome 2 (83 genes, 657 kb) and the rice chromosomes 9 (25 genes, 110 kb) and 7 (81 genes, 903 kb). Conserved genes are indicated with the same color code and linked with black lines. The SBP is indicated with a dotted black box. The dot plot illustrates the alignment of the 51-kb SBP region against itself.

Figure 3.

Impact of polyploidization of the genome structure. (A) Illustration of the synteny between Brachypodium chromosome 2 (3639 genes), sorghum chromosome 3 (4565 genes), rice chromosome 1 (5313 genes), maize chromosomes 3–8 (2873 and 1714 genes, respectively), and their ancestral relative (A5, 988 reordered genes in 96 blocks). Orthologous genes are linked with gray lines. Gene distribution (red curves) and percentage of conserved genes (blue bars) per megabase are provided for the Brachypodium, sorghum, and rice chromosomes (top). The duplication identified between maize chromosomes 3 and 8 as part of the recent tetraploidization are linked with gray lines and represent 499 orthologous relationships (bottom). (B) The ancestral duplication A11 and A12 is illustrated in modern grass species, i.e., rice (r11-r12), sorghum (s5-s8), and Brachypodium (b4-b4). The orthologous relationship between b4-r11-s5 (dark blue chromosomes at the top) and b4-r12-s8 (light blue chromosomes at the bottom) is illustrated with gray blocks. Paralogous gene distribution (bars) per megabase is provided for the three pairwise comparisons, i.e., b4-b4, r11-r12, s5-s8. The differential loss of duplicated genes in the subtelomeric region is illustrated with red bars within the gene distribution. The biased gene conversion (BGC) model for the observed differential loss of duplicated gene copies in the subtelomeric region is illustrated with the calculated nucleotide substitution rates (Ks values from 0–1, left scale) shown as green dots for the classes of paralogous couples highlighted with red bars. (C) Illustration of the synteny between wheat chromosome 1B and the ancestral relatives A5 and A7. The orthologous conserved genes are linked with colored lines. The height of wheat chromosome bin (from 1BS9 to 1BL3) is mentioned and is associated with the corresponding physical size (red dots), genetic size (blue dots), and number of conserved genes (gray bars), referring to the corresponding colored y-axis at the left of the figure. (D) Illustration of the synteny between barley chromosome 7 (displayed as chromosome heat map: blue, 5 markers; yellow, <5–10 markers; red, >10 markers in a 2-cM window) and the ancestral relatives A6 and A8. Orthologous conserved genes are linked with green (A6) and yellow (A8) lines. NCFs between A8 and A6 that occurred during the Triticeae genome paleo-history establish that A8 covers a 20-cM centromeric interval in the modern barley chromosome structure.

Figure 4.

Model for grass chromosome evolution and shuffling. The model begins with WGD (A), followed by centromeric breaks (B) or terminal breaks (C), interchromosomal break repair (D), and intra-CBR mechanisms (E, F), respectively, between nonhomologous (A) and homologous chromosomes (E) to explain the observed NCF and CI pattern (G) as well as repeat/TE, gene, and CO distribution in rice, maize, sorghum, and Brachypodium genomes, resulting from their paleo-history from their common AGK (H). Colored arrows represent the different alternative orientation of the double-strand break repair to explain the actual syntenic chromosome order and orientation observed among rice, Brachypodium, sorghum, maize.