Assessment of gene flow from a herbicide-resistant indica rice (Oryza sativa L.) to the Costa Rican weedy rice (Oryza sativa) in Tropical America: factors affecting hybridization rates and characterization of F1 hybrids

Abstract

Herbicide-resistant rice cultivars allow selective weed control. A glufosinate indica rice has been developed locally. However, there is concern about weedy rice becoming herbicide resistant through gene flow. Therefore, assessment of gene flow from indica rice cultivars to weedy rice is crucial in Tropical America. A field trial mimicking crop-weed growing patterns was established to assess the rate of hybridization between a Costa Rican glufosinate-resistant rice line (PPT-R) and 58 weedy rice accessions belonging to six weedy rice morphotypes. The effects of overlapping anthesis, morphotype, weedy accession/PPT-R percentage, and the particular weedy accession on hybridization rates were evaluated. Weedy rice accessions with short overlapping anthesis (4-9 days) had lower average hybridization rates (0.1%) than long anthesis overlapping (10-14 days) accessions (0.3%). Hybridization also varied according to weedy rice morphotype and accession. Sativa-like morphotypes (WM-020, WM-120) hybridized more readily than intermediate (WM-023, WM-073, WM-121) and rufipogon-like (WM-329) morphotypes. No hybrids were identified in 11 of the 58 accessions analyzed, 21 accessions had hybridization rates from 0.01% to 0.09%, 21 had rates from 0.1% to 0.9%, and 5 had frequencies from 1% to 2.3%. Another field trial was established to compare the weedy rice-PPT-R F(1) hybrids with their parental lines under noncompetitive conditions. F(1) hybrids had a greater phenotypic variation. They had positive heterosis for vegetative trait and reproductive potential (number of spikelets and panicle length) traits, but negative heterosis for seed set. This study demonstrated the complexity of factors affecting hybridization rates in Tropical America and suggested that the phenotype of F(1) hybrids facilitate their identification in the rice fields.