Understanding polyploid banana origins. A commentary on: ‘Unravelling the complex story of intergenomic recombination in ABB allotriploid bananas’

Abstract

This article comments on:

Alberto Cenci, Julie Sardos, Yann Hueber, Guillaume Martin, Catherine Breton, Nicolas Roux, Rony Swennen, Sebastien Christian Carpentier and Mathieu Rouard, Unravelling the complex story of intergenomic recombination in ABB allotriploid bananas, Annals of Botany, Volume 127, Issue 1, 01 January 2021, Pages 7–20, https://doi.org/10.1093/aob/mcaa032

While diploids are predominant among vertebrates, with a few exceptions, polyploids are pervasive among vascular plants, including both wild and cultivated species. Hybridization and whole genome duplication have long been considered the major evolutionary processes in angiosperms, leading sometimes to instant speciation and contributing to increased biodiversity (Alix et al., 2017). Two major forms of polyploids exist: auto- and allo-polyploids. The latter, in particular, are genetically and usually morphologically distinct from their diploid progenitors. Gametes that contain two sets of chromosomes play an important role in generating polyploidy in plants, and arise through meiotic defects (2n gametes) or in tetraploid plants (2n = 4x) during regular meiosis where haploid gametes (n) are actually numerically diploid (2x). Newly formed polyploids often possess novel physiological characteristics allowing them to settle in new habitats and exploit narrower ecological niches (Alix et al., 2017). They also have one important feature: the higher the level of ploidy the plant has, the bigger and heavier will be the seeds produced (Münzbergová, 2017); however, this may lead to numerous disturbances in endosperm development, as observed in synthetic allopolyploids (Tomaszewska and Kosina, 2018), and thus can affect seed germination. For breeders, allotriploids are of particular interest due to their increased heterozygosity and allelic diversity compared with their diploid progenitors, and their tendency to be sterile. Thus, they have been widely applied in the breeding of seedless fruit plants, including bananas, which is the subject of a paper in the current issue of Annals of Botany (Cenci et al., 2020).

Bananas are among the most numerously produced, consumed and traded fruits globally (FAO; www.fao.org), with more than 1000 varieties of bananas being grown for local consumption or commercial purposes throughout the tropics. In some African countries, banana is a staple food crop, and it plays an important role in food security, but pests and diseases are a growing threat to global banana production prompting steps towards genetic improvement of bananas (D’Hont et al., 2012). Therefore, it is important to accurately identify existing global genetic resources and recognize the evolutionary processes that have led to the emergence of species and new varieties (Heslop-Harrison and Schwarzacher, 2007).

Varietal improvement can be supported by molecular cytogenetic, genetic, genomic or bioinformatic methods through identification of the processes occurring when genomes come together in polyploids. To exploit variation in breeding, it is helpful to identify features of allopolyploidy that can be related to evolutionary adaptation, and can be identified as consequences of chromosomal rearrangements in the polyploid taxa (Gaeta et al., 2007). The genomes of many newly formed polyploids are unstable and may undergo exchanges. Intra- and inter-genomic rearrangements, such as large translocations, duplications, deletions or insertions, loss of chromatin or even whole chromosomes, and homoeologous exchanges affecting gene expression, may occur soon after polyploid formation, and give polyploids an adaptive and evolutionary advantage. In particular, changes in phenotype and fitness of a polyploid individual should guide the characterization of the genomes in diploid non-cultivated species to see how best to use them for germplasm enhancement.

Most banana varieties grown are autotriploids (AAA genomes) and allotriploids (AAB and ABB genomes) resulting from intra- and inter-specific hybridizations between Musa acuminata (A genome) and M. balbisiana (B genome). Three ancient rounds of independent whole genome duplication have been recognized in the Musa lineage, followed by gene loss and chromosome rearrangements which most probably allowed the diversification of species (D’Hont et al., 2012; Cenci et al., 2019). Large structural variations (translocations) in the genome of diploid M. acuminata subsp. malaccensis have been characterized, and the genomic changes involving homoeologous exchanges have been identified in polyploids. Here, Cenci et al. (2020) establish the genomic composition of important cooking bananas, and make substantial advances in identifying the events which led to the formation of modern allotriploid banana cultivars that so far had not been fully defined. They describe comprehensively the nature of the changes by using a combination of modern methods, recognizing the variation in genome structure in allotriploid bananas with a focus on the ABB group. Restriction site-associated DNA (RAD) markers were adapted for high-throughput sequencing to genotype single nucleotide polymorphisms (SNPs) in a representative group of 36 banana varieties with ABB genomes belonging to nine subgroups: Bluggoe, Monthan, Ney Mannan, KlueTeparod, Kalapua, Peyan, PisangAwak, Pelipita and Saba. The authors demonstrated with unprecedented accuracy that homoeologous exchanges between A and B genomes were common consequences of polyploidization among allotriploid bananas. The recognition of specific patterns of homoeologous exchanges in different banana varieties made it possible to determine the evolutionary paths that led to the diversification of current subgroups. It was important that diploid AB banana hybrids were included in the study, enabling the recognition of intergenomic chromatin exchange through intergenomic meiotic recombination which occurred before joining an additional genome in the triploid. The data presented by Cenci et al. (2020) allowed the conclusion that an unreduced gamete containing BB genomes originating from M. balbisiana played a significant role in the evolution of some allotriploid bananas. The routes which led to the creation of some of the ABB subgroups (Fig. 1) involve AB gametes originating from (i) the AABB allotetraploids, (ii) the AAB or ABB allotriploids or (iii) the AB hybrids. Some allotriploid plants may also be formed through elimination of chromosomes from allotetraploids (iv; see e.g. Kynast et al., 2002), but this is unlikely in the ABB banana varieties studied by Cenci et al. (2020).

Fig. 1.

Possible routes leading to the formation of ABB allotriploid plants. Diploid plants (grey box) give rise to haploid (reduced) gametes resulting in diploid offspring, or to diploid (unreduced) gametes resulting in tetraploid offspring. The allotriploid individual is usually formed in a two-step process involving backcrosses of: (i) AABB allotetraploid, (ii) BBBB autotetraploid or (iii) AB hybrids with the parental diploids. Elimination of chromosomes from allotetraploids (iv) can be another possible route leading to the creation of ABB allotriploids.

The results of Cenci et al. (2020) bring new insights into intergenomic recombination in triploids, and tell us about the multiple clusters of ABB banana cultivars and their taxonomic classification. Studies of intergenomic chromatin exchanges in different polyploid crops, such as that of Cenci et al. (2020) with allotriploid banana cultivars, provide essential information for understanding crop evolution, and can be extremely useful to develop crossing programmes to create new varieties with improved agronomic potential, and, more generally, can provide knowledge on common consequences of allotriploidization.

FUNDING

This project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie grant agreement No 844564 (to PT).