ABSTRACT: High acetylation of xylan in hardwoods decreases their value as biorefinery feedstocks. To counter this problem, we have constitutively suppressed RWA genes encoding acetyl-CoA transporters using the 35S promoter, or constitutively and wood-specifically (using the WP promoter) expressed fungal acetyl xylan esterases of families CE1 (AnAXE1) and CE5 (HjAXE), to reduce acetylation in hybrid aspen. All these transformations improved the saccharification of wood from greenhouse-grown trees. Here, we describe the chemical properties and saccharification potential of the resulting lines grown in a five-year field trial, and one type of them (WP:AnAXE1) in greenhouse conditions. Chemically, the lignocellulose of the field- and greenhouse-field-grown plants slightly differed, but the reductions in acetylation and saccharification improvement of engineered trees were largely maintained in the field. The main novel phenotypic observation in the field was higher lignification in lines with the WP promoter than those with the 35S promoter. Following growth in the field, saccharification glucose yields were higher from most transformed lines than from wild-type (WT) plants with no pretreatment, but there was no improvement in saccharification with acid pretreatment. Thus, acid pretreatment removes most recalcitrance caused by acetylation. We found a complex relationship between acetylation and glucose yields in saccharification without pretreatment, suggesting that other variables, for example, the acetylation pattern, affect recalcitrance. Bigger gains in glucose yields were observed in lines with the 35S promoter than in those with the WP promoter, possibly due to their lower lignin content. However, better lignocellulose saccharification of these lines was offset by a growth penalty and their glucose yield per tree was lower. In a comparison of the best lines with each construct, WP:AnAXE1 provided the highest glucose yield per tree from saccharification, with and without pretreatment, WP:HjAXE yields were similar to those of WT plants, and yields of lines with other constructs were lower. These results show that lignocellulose properties of field-grown trees can be improved by reducing cell wall acetylation using various approaches, but some affect productivity in the field. Thus, better understanding of molecular and physiological consequences of deacetylation is needed to obtain quantitatively better results.
