Biomass properties and enzyme-lignin interactions in the enzymatic cellulose degradation of hydrothermally pretreated lignocellulosic grass feedstocks

Abstract

Utilization of lignocellulosic biomass as renewable source of fuels, chemicals and materials under the biorefinery scheme offers sustainable option as compared to total reliance on using fossil fuels, which has been pointed as the culprit behind global warming and climate change. Hydrothermal pretreatment (HTP) of lignocellulosic biomass followed by enzymatic hydrolysis to produce monosaccharides have been developed mainly to produce fuel ethanol, yet the process is also meant to initiate the biorefining of biomass to produce platform monosaccharides and other chemicals. However, the process still suffers from the need to use high enzyme loading to produce reasonable product yield due to the inherent recalcitrance of lignocellulosic biomass. Research in optimizing the utilization lignocellulosic biomass is important to support the growing world’s population. This PhD study aimed to understand the factors that affect enzymatic cellulose degradation of lignocellulosic biomass that were processed using the latest technology. Biomass properties and enzyme-lignin interactions were of particular focus. Hydrothermally pretreated key grass biomass feedstocks (corn, Miscanthus, wheat) were investigated for their biomass properties to find correlation with the ensuing enzymatic digestibility using commercial cellulolytic enzyme mixture (Paper I). The corresponding lignin-rich residues (LRRs) from these pretreated biomass feedstocks were characterized for chemical and physical characteristics and studied for enzymelignin interaction and to assess the factors that retard enzymatic cellulose degradation (Paper II). A selected grass LRR from the previous work (wheat straw) and another LRR from hydrothermally pretreated softwood (spruce) were used to investigate the adsorption kinetics of monocomponent cellulases on lignin (Paper III). The tested grass biomass feedstocks had different digestibility after HTP at different severity levels (Paper I). However, these differences were not reflected in their bulk omposition, especially the extent of hemicellulose removal after HTP. Biomass wettability correlated well with digestibility and showed that the least digestible biomass had the lowest wettability after HTP. Analyses using attenuated total reflectance-Fourier transform infrared (ATR-FTIR) spectroscopy revealed that the least digestible biomass had the highest apparent surface abundance (ASA) of hemicellulose and lignin. Therefore, surface properties correlated better to the digestibility of hydrothermally pretreated grass biomass feedstocks compared to bulk composition (Paper I). The components of the commercial cellulolytic enzyme mixture adsorbed similarly on the isolated LRRs regardless of severity levels and biomass feedstocks, in agreement with minor chemical changes in lignin after HTP (Paper II). The LRRs also did not retard the enzymatic hydrolysis of model cellulose, suggesting reversible adsorption. The residual carbohydrates in the LRRs were not accessible to the enzymes and were not traceable to the surface of the LRRs using ATR-FTIR analysis. These data suggested that the enzymatic cellulose degradation was retarded by increasing presence of lignin in the surface of the biomass particles which can be affected by the physical properties of lignin. Thus lignin retards enzymatic cellulose degradation by acting as physical barrier rather than inducing non-productive adsorption (Paper II). The radiolabeled monocomponent cellulases had different binding affinity on the tested LRRs and fitted well with Langmuir model which assumes reversible binding (Paper III). Adsorption experiments with dilution at early and late time points revealed that the adsorption of the enzymes did not exhibit hysteresis. Kinetic modelling of the experiments showed that reversible adsorption behavior can fully explain the observed data, though pointed at extended reaction time and elevated temperature (40-50°C) the binding can turn irreversible. Furthermore, the adsorption parameters Kads and Bmax from both Langmuir fitting and kinetic modelling were in agreement. Simultaneous adsorption of different cellulases displayed competition and fitted well with Langmuir model. These observations gave compelling evidence of reversible binding of cellulases on lignin (Paper III). All in all the results above provided new understandings of biomass surface, lignin, and the interactions of cellulases with biomass, which should be considered and pursued further in order to advance the understanding in lignocellulosic biomass conversion.