Unraveling the sweetpotato’s genetic structure to speed up breeding

CIP scientists made key contributions to a pathbreaking study of two of the sweetpotato’s nearest relatives which helps to unravel the complex genetic structure of the sweetpotato itself. Mapping the genome of these two sweetpotato relatives, led by the Boyce Thompson Institute at Cornell University, provides high-quality references for the sweetpotato. The study also resequenced 16 genotypes of sweetpotato widely used in CIP’s African breeding programs, to identify genes that will help breed varieties with high vitamin A content

The sweetpotato has a large and complex genetic structure, with six sets of chromosomes and some 1.6 billion base pairs (half as many as humans). As an added twist, many genes are heterozygous. Because of this complexity, sweetpotato breeders have not had a good reference guide to the sweetpotato genome, making it difficult to apply genomic tools to enhance sweetpotato breeding. But a 2018 paper published in the journal Nature Communications sheds crucial light on the sweetpotato’s genes.

The sweetpotato (Ipomoea batatas), a member of the morning glory family, is closely related to the wild species Ipomoea trifida and Ipomoea triloba. These two species are similar to the sweetpotato in flower, vine and leaf, but without the edible roots. Another crucial difference is that both of the wild relatives have just two sets of chromosomes, like humans, but unlike the sweetpotato with its six sets of chromosomes.

Farmers planting OFSP.
H. Rutherford/CIP

For years, researchers have wondered if either of the wild species could be the ancestor of the sweetpotato. The Nature Communications article reveals that both may be. Most of the sweetpotato’s genes (84%) are also found in these two close relatives, but the sweetpotato shares 5.4% of its genes uniquely with Ipomoea trifida, and 3.4% only with Ipomoea triloba. This means that the sweetpotato probably has a mixed ancestry.

The Nature Communications paper reports on a complete analysis of the two wild Ipomoea species and of 16 sweetpotato varieties, a collection better known as MDP (Mwanga Diversity Panel), named after Robert Mwanga, the CIP sweetpotato breeder awarded the World Food Prize. Comparing the 16 varieties with the two wild Ipomoea genomes made it possible to find genetic markers which can help to locate genes associated with flesh color and other important traits.

To support this research, CIP’s role was to provide crucial populations and data to enhance the development of fully functional reference genomes that are representative of CIP’s target breeding programs. Among the populations provided by CIP are the I. trifida bi-parental mapping population that was used to anchor (assign chromosome positions using a genetic linkage map) to the reference genomes. Additionally, RNA from different experiments was provided to support annotation (identifying position and functions of genes) and access to the crucial 16 varieties.

Until now, most orange-fleshed sweetpotato (OFSP) varieties have been high in vitamin A but low in dry matter, which has contributed to limited adoption of new varieties. This research may help sweetpotato breeders to come up with sweetpotatoes that are high in vitamin A and starch that would provide a more nutritious and acceptable crop for African farmers and consumers. “However, CIP’s main role in supporting the tool development process is the interest to apply the tools, being the main applied research partner in this important collaboration. We have already started to apply the reference genomes and other novel analytical tools to answer our research questions such as understanding the negative association between starch and beta-carotene content, which is an important contributor to poor adoption of most OFSP in Africa, for example,” says Dorcus Gemenet, research scientist at CIP.

Good quality reference genomes and other data analytical tools can thus enhance practical applications of genomics in breeding programs. As remarked by Mwanga, “the high-quality reference genomes allow us to tease out the genetics of important traits that are difficult to target together and can provide us with molecular tools to facilitate breeding for several such desired traits.”