Comprehensive characterization of non-cellulosic recalcitrant cell wall carbohydrates in unhydrolyzed solids from AFEX-pretreated corn stover

BACKGROUND: Inefficient carbohydrate conversion has been an unsolved problem for various lignocellulosic biomass pretreatment technologies, including AFEX, dilute acid, and ionic liquid pretreatments. Previous work has shown 22% of total carbohydrates are typically unconverted, remaining as soluble or insoluble oligomers after hydrolysis (72 h) with excess commercial enzyme loading (20 mg enzymes/g biomass). Nearly one third (7 out of 22%) of these total unconverted carbohydrates are present in unhydrolyzed solid (UHS) residues. The presence of these unconverted carbohydrates leads to a considerable sugar yield loss, which negatively impacts the overall economics of the biorefinery. Current commercial enzyme cocktails are not effective to digest specific cross-linkages in plant cell wall glycans, especially some of those present in hemicelluloses and pectins. Thus, obtaining information about the most recalcitrant non-cellulosic glycan cross-linkages becomes a key study to rationally improve commercial enzyme cocktails, by supplementing the required enzyme activities for hydrolyzing those unconverted glycans. RESULTS: In this work, cell wall glycans that could not be enzymatically converted to monomeric sugars from AFEX-pretreated corn stover (CS) were characterized using compositional analysis and glycome profiling tools. The pretreated CS was hydrolyzed using commercial enzyme mixtures comprising cellulase and hemicellulase at 7% glucan loading (~20% solid loading). The carbohydrates present in UHS and liquid hydrolysate were evaluated over a time period of 168 h enzymatic hydrolysis. Cell wall glycan-specific monoclonal antibodies (mAbs) were used to characterize the type and abundance of non-cellulosic polysaccharides present in UHS over the course of enzymatic hydrolysis. 4-O-methyl-d-glucuronic acid-substituted xylan and pectic-arabinogalactan were found to be the most abundant epitopes recognized by mAbs in UHS and liquid hydrolysate, suggesting that the commercial enzyme cocktails used in this work are unable to effectively target those substituted polysaccharide residues. CONCLUSION: To our knowledge, this is the first report using glycome profiling as a tool to dynamically monitor recalcitrant cell wall carbohydrates during the course of enzymatic hydrolysis. Glycome profiling of UHS and liquid hydrolysates unveiled some of the glycans that are not cleaved and enriched after enzyme hydrolysis. The major polysaccharides include 4-O-methyl-d-glucuronic acid-substituted xylan and pectic-arabinogalactan, suggesting that enzymes with glucuronidase and arabinofuranosidase activities are required to maximize monomeric sugar yields. This methodology provides a rapid tool to assist in developing new enzyme cocktails, by supplementing the existing cocktails with the required enzyme activities for achieving complete deconstruction of pretreated biomass in the future. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s13068-017-0757-5) contains supplementary material, which is available to authorized users.