In grasses, residues homologous to residues Ile-1,781 and Ile-2,041 in the carboxyl-transferase (CT) domain of the chloroplastic acetyl-coenzyme A (CoA) carboxylase (ACCase) from the grass weed black-grass (Alopecurus myosuroides [Huds.]) are critical determinants for sensitivity to two classes of ACCase inhibitors, aryloxyphenoxypropionates (APPs) and cyclohexanediones. Using natural mutants of black-grass, we demonstrated through a molecular, biological, and biochemical approach that residues Trp-2,027, Asp-2,078, and Gly-2,096 are also involved in sensitivity to ACCase inhibitors. In addition, residues Trp-2,027 and Asp-2,078 are very likely involved in CT activity. Using three-dimensional modeling, we found that the side chains of the five residues are adjacent, located at the surface of the inside of the cavity of the CT active site, in the vicinity of the binding site for APPs. Residues 1,781 and 2,078 are involved in sensitivity to both APPs and cyclohexanediones, whereas residues 2,027, 2,041, and 2,096 are involved in sensitivity to APPs only. This suggests that the binding sites for these two classes of compounds are overlapping, although distinct. Comparison of three-dimensional models for black-grass wild-type and mutant CTs and for CTs from organisms with contrasted sensitivity to ACCase inhibitors suggested that inhibitors fitting into the cavity of the CT active site of the chloroplastic ACCase from grasses to reach their active sites may be tight. The three-dimensional shape of this cavity is thus likely of high importance for the efficacy of ACCase inhibitors.

Summary: Pleiotropic effects associated with genes endowing resistance to herbicides are generally predicted to reduce plant fitness. Quantifying these effects is necessary to develop management strategies against herbicide‐resistant weeds. We assessed the pleiotropic effects associated with three mutant alleles of the herbicide target enzyme acetyl‐coenzyme A carboxylase (ACCase) on plant growth and seed production in black‐grass Alopecurus myosuroides. In each of two field experiments, black‐grass populations segregating for Leu‐1781 (five populations), Asn‐2041 (three populations) or Gly‐2078 (two populations) ACCase were produced to obtain several distinct, homogenized genetic backgrounds and to permit reliable comparisons among wild‐type, heterozygous and homozygous mutant ACCase plants grown in competition with a wheat crop. No significant differences from wild‐type plants in vegetative biomass, height and seed production were observed in Leu‐1781 or Asn‐2041 ACCase plants. Over both experiments, homozygous Gly‐2078 ACCase plants displayed a significant reduction in biomass (42%), height (6%) and seed production (36%). Reduction varied with the segregating population and between field experiments. Synthesis and applications. Our work illustrates the variation in fitness cost depending on the resistance gene, the plant genetic background and the environment. This underlines the necessity to identify the resistance gene(s) present in a weed population before designing resistance‐management strategies. Competitive crops should be effective against Gly‐2078 ACCase plants. The effect of resistant ACCase alleles on seed survival and germination needs to be studied in order to develop cultural practices creating or maximizing fitness costs in resistant plants. However, the variation of fitness cost with the environment and the weed population, as well as the likely absence of fitness cost associated with resistance genes such as Leu‐1781 ACCase, renders the success of specific cultural control practices uncertain. A solution could be a weed control programme maximizing the diversity of cultural practices and including anti‐resistance cultural practices to avoid or reduce selection for resistant ACCase alleles.