Cell Wall Acetylation in Hybrid Aspen Affects Field Performance, Foliar Phenolic Composition and Resistance to Biological Stress Factors in a Construct-Dependent Fashion

Abstract

The production of biofuels and "green" chemicals from the lignocellulose of fast-growing hardwood species is hampered by extensive acetylation of xylan. Different strategies have been implemented to reduce xylan acetylation, resulting in transgenic plants that show good growth in the greenhouse, improved saccharification and fermentation, but the field performance of such plants has not yet been reported. The aim of this study was to evaluate the impact of reduced acetylation on field productivity and identify the best strategies for decreasing acetylation. Growth and biological stress data were evaluated for 18 hybrid aspen lines with 10-20% reductions in the cell wall acetyl content from a five year field experiment in Southern Sweden. The reduction in acetyl content was achieved either by suppressing the process of acetylation in the Golgi by reducing expression of REDUCED WALL ACETYLATION (RWA) genes, or by post-synthetic acetyl removal by fungal acetyl xylan esterases (AXEs) from two different families, CE1 and CE5, targeting them to cell walls. Transgene expression was regulated by either a constitutive promoter (35S) or a wood-specific promoter (WP). For the majority of transgenic lines, growth was either similar to that in WT and transgenic control (WP:GUS) plants, or slightly reduced. The slight reduction was observed in the AXE-expressing lines regulated by the 35S promoter, not those with the WP promoter which limits expression to cells developing secondary walls. Expressing AXEs regulated by the 35S promoter resulted in increased foliar arthropod chewing, and altered condensed tannins and salicinoid phenolic glucosides (SPGs) profiles. Greater growth inhibition was observed in the case of CE5 than with CE1 AXE, and it was associated with increased foliar necrosis and distinct SPG profiles, suggesting that CE5 AXE could be recognized by the pathogen-associated molecular pattern system. For each of three different constructs, there was a line with dwarfism and growth abnormalities, suggesting random genetic/epigenetic changes. This high frequency of dwarfism (17%) is suggestive of a link between acetyl metabolism and chromatin function. These data represent the first evaluation of acetyl-reduced plants from the field, indicating some possible pitfalls, and identifying the best strategies, when developing highly productive acetyl-reduced feedstocks.

Keywords: AnAXE1; HjAXE; Populus tremula × tremuloides; biotic resistance; condensed tannins; field trial; salicinoid phenolic glucosides; transgenic trees.

FIGURE 1

Field testing revealed striking phenotypes in three out of 20 lines tested, effects which could not be related to transgenes. Overview of the field trial in July 2017 (fourth year) (A) and the corresponding data for anomalous lines: stem volume (B), apical dominance (C) and mortality within the lines (D). Lines marked in red showed aberrant morphology compared to other lines with the same construct and to WT. Line 11 with the WP:HjAXE construct had higher mortality than all other lines, and exhibited a variegated phenotype (E) not seen in other lines carrying this construct. Scale bar in E – 2 cm. Data in panels (B) and (C) are means ± SE.

FIGURE 2

Effects of different types of genetic modification on growth of transgenic lines during a four-year field trial. Height (A), diameter (B), and stem volume (C). Means and SE, stars indicate means significantly different from WT, post-ANOVA contrast, p ≤ 0.05). P-value indicates significance of the promoter effect based on a two-way ANOVA (Supplementary Table S6).

FIGURE 3

Effects of different types of genetic modification on leaf traits. Leaf dry weight with representative images of leaves, size bar = 1 cm (A). Leaf chlorophyll index (B). Means and SE, stars indicate means significantly different from WT, post-ANOVA contrast, p ≤ 0.05). P-value indicates significance of the promoter effect based on a two-way ANOVA (Supplementary Table S6) and synthetic vs. post-synthetic xylan modification (Supplementary Table S7).

FIGURE 4

Activity of wood-specific promoter (WP) in trees grown in the field. Histochemical β-glucuronidase analysis of a branch with one-year old cambium of WP:GUS trees (lines 25 and 27) in the fourth growing season in July, during the active wood production period. X – secondary xylem; P – secondary phloem, PF – phloem fibers; C-vascular cambium. Activity is seen in cells depositing secondary cell walls (marked with brackets). Scale bar = 100 μm.

FIGURE 5

Effects of different types of genetic modification on biotic stress responses. Instances of leaf damage recorded in the 2017 survey that showed significant effects of “construct.” (A) Necrosis and rust. Constructs producing significantly different distributions are marked with *. Details of statistical analysis are provided in Supplementary Table S5. (B) Chewing damage. (C) Condensed tannin contents. Data in B and C are means and SE, stars indicate means significantly different from WT (post-ANOVA contrast, p ≤ 0.05). P-value indicates significance of the transgene or promoter effect based on a two-way ANOVA (Supplementary Table S6) and synthetic vs. post-synthetic xylan modification strategy (Supplementary Table S7).