Reproductive system and molecular phylogenetic relationships of Fonio Millets (Digitaria spp., Poaceae) with some polyploid wild relatives

2010 
Fonio millets (Digitaria exilis, D. iburua) are minor but important indigenous cereals in the semiarid areas of West-Africa. Recent interest in breeding strategies for these crops requires a better understanding of their biology and genetics. Amplified Fragment Length Polymorphism (AFLP) markers were employed to assess the phylogenetic relationships among cultivated fonio species and some polyploid wild relatives and examine proposed hypotheses on fonio ancestry. The AFLP analysis was found quite suitable for identifying each species. A very strong genetic affinity (over 92% similarity) was detected between the wild D. longiflora and D. ternata and the cultivated D. exilis and D. iburua, respectively. These data provided additional molecular evidence supporting the previous view of direct domestication of fonio millets from these two wild species. High genetic divergences were expectedly found between fonio species and the other taxonomically distant Digitaria taxa investigated. The results also revealed D. ciliaris and D. sanguinalis as separate species sharing close ancestry. Selfing experiments and subsequent progeny analyses using three isozymes supplemented by AFLPs were further conducted to determine the reproductive system in fonio millets. The results revealed apomixis as absolute mode of reproduction of these crops, except D. exilis in which 2% residual sexuality was detected. Additional data documented on seed set and pollen viability suggested that apomixis in fonio would be of pseudogamous type. The data also revealed fonio crops as highly self-compatible and of allopolyploid origin. This study adds new information about the reproductive system and the evolution of fonio, contributing to the knowledge on their biology, and thus providing useful subsides for future genetic improvement of these valuable crops.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    60
    References
    15
    Citations
    NaN
    KQI
    []