Reaction wood - a key cause of variation in cell wall recalcitrance in willow

Abstract

Background: The recalcitrance of lignocellulosic cell wall biomass to deconstruction varies greatly in angiosperms, yet the source of this variation remains unclear. Here, in eight genotypes of short rotation coppice willow (Salix sp.) variability of the reaction wood (RW) response and the impact of this variation on cell wall recalcitrance to enzymatic saccharification was considered.

Results: A pot trial was designed to test if the 'RW response' varies between willow genotypes and contributes to the differences observed in cell wall recalcitrance to enzymatic saccharification in field-grown trees. Biomass composition was measured via wet chemistry and used with glucose release yields from enzymatic saccharification to determine cell wall recalcitrance. The levels of glucose release found for pot-grown control trees showed no significant correlation with glucose release from mature field-grown trees. However, when a RW phenotype was induced in pot-grown trees, glucose release was strongly correlated with that for mature field-grown trees. Field studies revealed a 5-fold increase in glucose release from a genotype grown at a site exposed to high wind speeds (a potentially high RW inducing environment) when compared with the same genotype grown at a more sheltered site.

Conclusions: Our findings provide evidence for a new concept concerning variation in the recalcitrance to enzymatic hydrolysis of the stem biomass of different, field-grown willow genotypes (and potentially other angiosperms). Specifically, that genotypic differences in the ability to produce a response to RW inducing conditions (a 'RW response') indicate that this RW response is a primary determinant of the variation observed in cell wall glucan accessibility. The identification of the importance of this RW response trait in willows, is likely to be valuable in selective breeding strategies in willow (and other angiosperm) biofuel crops and, with further work to dissect the nature of RW variation, could provide novel targets for genetic modification for improved biofuel feedstocks.

Figure 1

A Illustrations depicting the traditional notion of reaction wood. Top: a single stem bent away from the vertical, Bottom: a transverse section showing the tension wood region more darkly shaded. B Images displaying reaction wood in field-grown willow. Top: Photograph of mature willow stems grown in a UK field trial. Bottom: Midpoint 20-μm transverse section of a single stem from a mature field-grown willow tree. Stained in 1% Chlorazol Black E in methoxyethanol (black – binds specifically to the gelatinous layer within the G-fibres of tension wood [21]) and 1% aqueous Safranin O (red – binds to the secondary cell wall in a non-specific manner). Scale bar = 5 mm.

Figure 2

Control and Reaction Wood induced potgrown trees of eight genotypes.A Glucan composition expressed as a percentage of dry matter (DM). B Lignin composition expressed as a percentage of DM. C Glucose yields from enzymatic saccharification presented as grams of glucose released per gram of glucan present in the biomass. Error bars represent standard error (n = 3 trees). Full mass closed compositional tables are available in supplementary information. D Midpoint 20-μm transverse sections of a single stem from pot-grown genotypes ‘Shrubby’ and ‘K8-428’. Stained in 1% Chlorazol Black E in methoxyethanol (black – binds specifically to the gelatinous layer within the G-fibres of tension wood [21]) Scale bar = 5 mm. * Significant difference (t-test, p < 0.05).

Figure 3

Correlations of glucose yields from enzymatic saccharification for eightSalixgenotypes. Glucan accessibility from mature field-grown (Rothamsted Research site – RRes) trees correlated against glucan accessibility from: A control pot-grown trees and B reaction wood induced pot-grown trees. Glucan accessibility expressed as grams of glucose release per gram of glucan present in the biomass. Correlation coefficients and significance level displayed. Error bars represent standard error (n = 3 trees).

Figure 4

Mature field-grown trees of eight genotypes grown at the Rothamsted Research (RRes) and Orkney sites.A Glucan composition expressed as a percentage of dry matter (DM). B Lignin composition expressed as a percentage of DM. C Glucose yields from enzymatic saccharification presented as grams of glucose release per gram of glucan present in the biomass. Error bars represent standard error (n = 3 trees). Full mass closed compositional tables are available in supplementary information. * Significant difference (t-test, p < 0.05).

Figure 5

Comparison of enzymatic saccharification yields of the three genotypes present in the pot trial and the Rothamsted Research (RRes) and Orkney field sites. Yields are presented as grams of glucose release per gram of biomass and so encompass variation in both glucan content and glucan accessibility. Low reaction wood inducing (RWI) conditions = control trees (pot) and RRes site (field). High RWI conditions = RW induced trees (pot) and Orkney site (field). Error bars represent standard error (n = 3 trees). * Significant difference (t-test, p < 0.05).