Genetic variation and inheritance characteristics for carbon isotope discrimination in alfalfa.

Abstract

The negative correlation between carbon isotope discrimination and water-use efficiency in C3 species, including alfalfa (Medicago sativa L.), suggests that carbon isotope discrimination that might be useful in the selection of alfalfa cultivars that use water more efficiently. We initiated field experiments with alfalfa in northern Utah to determine genetic variation for carbon isotope discrimination within representative breeding populations, the effect of drought on carbon isotope discrimination, magnitudes of heritability for carbon isotope discimination, genetic regulation of carbon isotope discrimination, and how carbon isotope discrimination differs among plant parts. In an experiment conducted under a rainout shelter facility equipped with a line-source sprinkler system, genetic variability for carbon isotope discrimination was not detected in 15 clones each from the NC-83-1 germplasm and 'Spredor 2' cultivar. In another experiment with 25 clones from the NC-83-1 germplasm, there was significant (P < 0.01) genetic variation for carbon isotope discrimination with a range of 1.6 per thousand, and broad-sense heritabilities exceeded 0.80. In a field trial with 78 cultivars and elite breeding lines, significant genetic variation for carbon isotope discrimination was observed, although the range for carbon isotope discrimination was only 0.8 per thousand. We also detected significant genetic variation for carbon isotope discrimination in a diallel experiment with 196 crosses from 14 parent clones from NC-83-1. Furthermore, general combining ability was significant, but specific combining ability and reciprocal effects were not, indicating that standard breeding techniques could be used to alter carbon isotope discrimination response in alfalfa. Plant parts differed significantly for carbon isotope discrimination with stems having the lowest value (18.7 per thousand) followed by the entire shoot (19.0 per thousand), upper leaves (19.4 per thousand), and bottom leaves (20.2 per thouand). The lack of significant statistical interactions among plant parts suggested that any plant part could be sampled to determine carbon isotope discrimination. The results from these experiments indicated that promise exists for using carbon isotope discrimination to improve water-use efficiency in alfalfa; however, use of more diverse germplasm may be necessary to expand opportunities for selection in North American alfalfa germplasm.