Characterisation and enzymic degradation of non-starch polysccharides in lignocellulosic by-products : a study on sunflower meal and palm-kernel meal

Non-starch polysaccharides (NSP) constitute a potentially valuable part of plant by- products deriving from the food and agricultural industries. Their use for various applications (fuel, feed, food) requires the degradation and modification of the complex plant materials. This can be achieved by enzymatic processes which, in comparison with chemical or physical methods, are regarded as energy-saving and non-polluting. However, a major disadvantage of enzymic processes often is their low effectivity and consequently high costs.The investigations described in this thesis were conducted to understand the reasons for the low susceptibility to enzymic hydrolysis of such by-products and, in particular, of their non-starch polysaccharides, and to find out whether and how the efficacy of enzymic treatment could be enhanced. The studies should provide information necessary for the development of polysaccharidase-preparations, tailored for the use in different applications.Sunflower ( <em>Helianthus annuus</em> L.) meal and palm-kernel ( <em>Elaeis guineensis</em> Jacq) meal, by-products from the production of edible oil and used as animal feed compounds, were chosen for our studies.In <strong>chapter 1</strong> an introduction is given to the macroscopic and microscopic structure of the raw materials, to plant cell walls and their constituent polymers. The biodegradation of cell walls and its limitations are briefly reviewed and major non-starch polysaccharide degrading enzymes are summarised. Chapter 1 closes with an outline of the thesis.For a detailed study of type and structure of the non-starch polysaccharides, cell wall materials (CWM) were prepared from the meals by enzymatic digestion of protein and removal of small amounts of buffer-extractable material ( <strong>chapter 2</strong> ). The resulting CWM's were enriched in NSP (55% of sunflower CWM and 75% of palm-kernel CWM) and had a high lignin content. Two different chemical methods, sequential extraction with alkali and sodium chlorite and treatment with 4-methylmorpholine-N-oxide (MMNO) were tested to extract all constituent polysaccharides. Almost complete dissolution could be achieved by a combination of both methods, but the recovery of sugars, especially during MMNO treatment, was low. From the sugar composition of polysaccharide fractions, obtained by sequential chemical extraction, a tentative identification of major polysaccharides was achieved. Their distribution in different botanical fractions of the meals could be deduced by comparison with data from literature (sunflower) or by own experiments (palm-kernel).The polysaccharide extracts of different purity were further fractionated by graded precipitation with ethanol, size-exclusion or adsorption-chromatography. By determination of the sugar- and glycosidic linkage composition of extracts, (partially) purified subfractions and intact cell wall materials, the identification, partial characterisation and quantification of major non-starch polysaccharides were achieved ( <strong>chapter 3</strong> ). In sunflower meal, cellulose (42% of NSP), pectic polysaccharides (24%) and (4-O-methyl)-glucuronoxylans (24%) with about 10% glucuronosyl-substitution were major constituents. Minor amounts of (gluco)mannans (5%) and fucoxyloglucans (4.5%) were also identified. Major polysaccharides in palm-kernel meal were mannans (78% of NSP) with very low degrees of galactose-substitution and of apparently small molecular size (DP 12 to 14), and cellulose (12%). Arabinoxylans (3%) and (4- <em>O</em> -methyl)-glucuronoxylans (3%), deriving from the endocarp fraction of the meal, were present in low amounts in this monocotyledenous material.For a study of the enzymic hydrolysis of the cell wall materials ( <strong>chapter 4</strong> ), three multi-component enzyme preparations were chosen. Solubilisation occurred as a bi-phasic process with high reaction velocities in the first stage of the incubation and only slow progress during extended incubation up to 72h. The solubilisation could markedly be improved by reduction in particle size; partial delignification or increasing enzyme concentration, however, had almost no effect. Maximally 30% of NSP in sunflower meal and 50% in palm-kernel meal could be solubilised from the finely milled CWM's. Although the composition of the enzyme preparations was found to influence the type of reaction products, the extent of their release and, as observed by transmission electron microscopy, the site of enzymic attack in different cell wall layers, our results suggested that substrate accessibility was the major factor limiting enzymic hydrolysis.A detailed study of the reaction products obtained by incubation with the crude enzyme preparations or fractions thereof (prepared by anion-exchange chromatography) revealed, that pectic compounds and mannose-containing polysaccharides in sunflower CWM were readily degradable ( <strong>chapter 5</strong> ). The hydrolysis of mannans in palm-kernel CWM varied from 20% to 50%. In both CWM's, xylans and cellulose were most resistant to hydrolysis. The results indicate the preferential degradation of parenchyma and endosperm tissues and the resistance of hull and endocarp fractions to enzymic hydrolysis. The reaction products formed during all stages of the treatment were of small oligomeric and monomeric size.The contribution of different enzyme activities to the total solubilisation achieved by the heterogeneous enzyme preparations was studied with (partially) purified subfractions which were prepared by various chromatographic techniques from the crude preparations, and with highly purified enzymes from other microbial sources ( <strong>chapter 6</strong> ). In general, the effect of these purified enzyme fractions was low (solubilisation of NSP: 1 % to 5 %). Supplementation of main enzyme fractions with pectolytic, cellulolytic or mannanolytic subfractions did only slightly enhance the total solubilisation. Synergistic action was observed between glucanases and mannanases in palm-kernel incubations and between arabinanases, polygalacturonases and rhamnogalacturonan-degrading enzyme fractions in the hydrolysis of pectic polysaccharides in sunflower CWM. The enzymic hydrolysis of (4- <em>O</em> -methyl)-glucuronoxylans was studied <em>in-situ</em> and with the isolated polysaccharide. The results indicated that the resistance of the xylans to enzymic degradation is not only due to their interlinkage with other polymers and location in the cell wall but also to their primary structure.In <strong>chapter 7</strong> , important aspects concerning the approach and the methodology used are discussed. Implications arising for different fields of application are shown and suggestions for the formulation of enzyme preparations, which merit further research, are made.