A project undertaken at Australia National University, and supervised by Allan Lohe
With growing pressure from climate change and increasing world population, the resilience, sustainability and yields of our crop species must be continually improved to address these challenges. Many of the mainstay crops of the world’s food production are allopolyploids, such as wheat, canola and soybean. Allopolyploids are formed by hybridisation between two different but related species of plants followed by chromosome doubling. For example, the allotetraploid canola which originated in nature about 7,000 years ago, contains two diploid genomes, one from Brassica rapa and the other from B. oleracea. The germplasm of allopolyploids originating in nature like canola also shows a greatly improved response in plant breeding, often surpassing both parents. However, because these allopolyploids arose naturally thousands of years ago in specific regions, they can exhibit limited stress/disease resilience despite having the desired agronomic traits of greater vigour and yield.
It would be highly advantageous in crop breeding programs to resynthesise useful allopolyploids efficiently from different cultivars, for example species with desirable biotic or abiotic traits, or species from different climates or with specific stress-resistance genes. It might also be possible to generate novel allopolyploids using parents from different related species. This would provide a rapid, non-GMO, breeding strategy of introducing desired genes. Unfortunately, a major roadblock in replicating such wide-crosses in the laboratory (synthetically) is that new allopolyploids are sterile or near-sterile. These synthetic allopolyploids produce mostly shrivelled seed and consequently are not useful in agriculture without time-consuming selection for plump seed over many generations.
In this project we are investigating the biological basis for sterility and chromosomal instability of synthetic allopolyploids, apparently due to defective gametes produced from anomalies in meiosis. To achieve error-free meiosis, chromosomes must pair and segregate correctly in order to halve the chromosome number of the parent in the gametes. In diploids, pairing in meiosis involves homologous chromosomes (that is, identical chromosomes with one inherited from each parent) recognising each other and associating to form a bivalent, which describes the two closely-associated homologous chromosomes when paired. Once a bivalent is formed, homologous chromosomes can segregate and thereby halve their chromosome number in the gametes after meiosis.
Allopolyploids face a different situation at meiosis because they contain two sets of closely-related chromosomes, usually one diploid set from each of the parental species. A chromosome in an allopolyploid has the choice of pairing with its true homologue (a partner from the same parental species) or with a heterologous chromosome (a partner from the other parental species).
In naturally-occurring allopolyploids chromosome pairing in meiosis always occurs in a diploid-like manner, that is between homologous chromosomes, to form bivalents. Heterologous chromosome pairing occurs rarely, if at all. In contrast, synthetic allopolyploids often exhibit heterologous pairing at meiosis, resulting in multivalents (three or more chromosomes pairing with each other) and lagging chromosomes, leading to unbalanced gametes and shrivelled seed. In naturally occurring allopolyploids, loci stabilising meiosis have arisen or been selected for to overcome this problem. For example, homologous chromosome pairing in allohexaploid wheat is promoted by the Ph1 locus. Functionally analogous loci have been identified in canola and Arabidopsis suecica, both naturally-arisen allotetraploids. It is not known how these loci arose but it may have occurred over many generations by selection for plump seed, indirectly selecting for loci stabilising meiosis. Alternatively, such loci might have appeared spontaneously upon allopolyploid generation.
A gap in our knowledge is whether existing loci promoting homologue pairing in natural allopolyploids might also stabilise newly-synthesised allopolyploids when introduced genetically. In proof-of-concept experiments we found that a synthetic Arabidopsis allotetraploid hybrid was near-sterile, but when crossed with a natural Arabidopsis allotetraploid some fertility could be rescued, creating a new species. Backcrossing with the natural allotetraploid further improved seed yield. This suggests the natural allotetraploid, which contains loci promoting correct chromosome pairing and recombination at meiosis, can restore fertility to the synthetic allotetraploid.
We are monitoring meiosis cytologically in these plants to determine how it progresses, and characterising genome stability of this new species at a molecular level. Next, we will translate our findings into agriculturally important crop plants by resynthesising novel allopolyploids from wild progenitors and introducing meiosis-stabilising loci genetically, potentially generating fully fertile and genomically stable crop plants with better agronomic traits.