Targets and tools for optimizing lignocellulosic biomass quality of miscanthus

Miscanthus is a perennial energy grass characterized by a high productivity and resource-use efficiency, making it an ideal biomass feedstock for the production of cellulosic biofuels and a wide range of other biobased value-chains. However, the large-scale commercialization of converting biomass into cellulosic biofuel is hindered by our inability to efficiently deconstruct the plant cell wall. The plant cell wall is a complex and dynamic structure and its components are extensively cross-linked into an unyielding matrix. The production of biofuel depends on the extraction, hydrolysis and fermentation of cell wall polysaccharides, which currently requires energetically and chemically intensive processing operations that negatively affect the economic viability and sustainability of the industry. To address this challenge it is envisioned that the bioenergy feedstocks can be compositionally tailored to increase the accessibility and extractability of cell wall polysaccharides, which would allow a more efficient conversion of biomass into biofuel under milder processing conditions. Extensive phenotypic and genetic diversity in cell wall composition and conversion efficiency was observed in different miscanthus species, including <em>M. sinensis</em>, <em>M. sacchariflorus</em> and interspecific hybrids between these two species. In multiple experiments a twofold increase in the release of fermentable sugars was observed in ‘high quality’ accessions compared to ‘low quality’ accessions. The exhaustive characterization of eight highly diverse <em>M. sinensis</em> genotypes revealed novel and distinct breeding targets for different bioenergy conversion routes. The key traits that contributed favourably to the conversion efficiency of biomass into biofuel were a high content of hemicellulosic polysaccharides, extensive cross-linking of hemicellulosic polysaccharides (revealed by a high content of <em>trans-</em>ferulic acids and a high ratio of arabinose-to-xylose), a low lignin content and extensive incorporation of <em>para</em>-coumaric acid into the lignin polymer. Lignin is widely recognized as one of the key factors conveying recalcitrance against enzymatic deconstruction of the cell wall. The incorporation of <em>para</em>-coumaric acid into the lignin polymer is hypothesized to make lignin more easily degradable during alkaline pretreatment, one of the most widely applied processing methods that is used to pretreat biomass prior to enzymatic hydrolysis. Previous studies have shown that reducing lignin content is often implicated in reduced resistance of plants to lodging. We hypothesize that extensively cross-linked hemicellulosic polysaccharides may fulfil a similar function in supporting cell wall structural rigidity and increasing the content of hemicellulosic polysaccharides may be a way to reduce lignin content without adversely affecting cell wall rigidity. This strategy can be used to improve biomass quality for biobased applications, as hemicellulosic polysaccharides are more easily degradable during industrial processing than lignin. Furthermore, hemicellulosic polysaccharides adhere to cellulose, which negatively affects the level of cellulose crystallinity. Crystalline cellulose is harder to degrade than its more amorphous form. Therefore the reduction of cellulose crystallinity is another mechanism through which increasing the content of hemicellulosic polysaccharides positively contributes to cell wall degradability. These results provided new insights into the traits that may be targeted to improve the quality of lignocellulose feedstocks. However, evaluation of complex biochemical traits for selection purposes is hindered by the fact that their accurate quantification is a costly, lengthy and laborious procedure. To overcome these limitations an accurate and high-throughput method was developed based on near-infrared spectroscopy. Through extensive calibration we developed accurate prediction models for a wide range of biomass quality characteristics, which may be readily implemented as a phenotyping tool for selection purposes. Additionally, progress through breeding may substantially be improved by marker-assisted selection, which will reduce the need for the evaluation of genotype performance in multi-year field trials. To this end, a biparental <em>M. sinensis</em> mapping population of 186 individuals was developed and genotyped using a genotyping-by-sequencing approach. A total of 564 short-sequence markers were used to construct a new <em>M. sinensis</em> genetic map. Cell wall composition and conversion efficiency were observed to be highly heritable and quantitatively inherited properties. This is the first genetic study in miscanthus to map quantitative trait loci (QTLs) for biomass quality properties and is a first step towards the application of marker-assisted selection for biomass quality properties. Through the evaluation of a diverse set of miscanthus genotypes in multiple locations we demonstrated that in addition to genotypic variation, growing conditions may have a substantial influence on cell wall composition and conversion efficiency. While further research is needed to identify which specific environmental parameters are responsible for the observed effects, these results clearly indicate that the environmental influence on biomass quality needs to be taken into account in order to match genotype, location and end-use of miscanthus as a lignocellulose feedstock. Moreover, significant genotype-by-environment interaction effects were observed for cell wall composition and conversion efficiency, indicating variation in environmental sensitivity across genotypes. Although the magnitude of the genotypic differences was small in comparison to genotype and environmental main effects, this affected the ranking of accession across environments. Stability analysis indicated some stable accessions performed relatively across diverse locations. In addition to trialing miscanthus in diverse locations, we also evaluated miscanthus biomass quality under drought conditions for a number of reasons: 1) drought stress is linked to a differential expression of cell wall biosynthesis genes, 2) incidence of drought events is increasing due to climate change, 3) irrigation is likely to be uneconomical during the cultivation of miscanthus and 4) miscanthus has many characteristics that make it a crop with a good potential for cultivation on marginal soils, where abiotic stresses such as drought may prevail. Drought stress was shown to result in a large reduction in cell wall and cellulose content and a substantial increase in hemicellulosic polysaccharides and cellulose conversion rates. We hypothesized that the reduction in cellulose content was due to an increase in the production of osmolytes, which are well-known for their role in plant protection against drought. The results indicated that drought stress had a positive effect on the cell wall degradability of miscanthus biomass. Overall the compendium of knowledge generated within the framework of this thesis provided insights into the variation in biomass quality properties in miscanthus, increased our understanding of the molecular, genetic and environmental factors influencing its conversion efficiency into biofuel and provided tools to exploit these factors to expand the use of miscanthus as a lignocellulose feedstock.