Skip to main content

Valorisation of cereal by-products: a biorefinery approach

Time: Fri 2022-03-04 10.00

Location: Kollegiesalen, Brinellvägen 8, Stockholm

Video link: https://kth-se.zoom.us/webinar/register/WN_f7jnKaXhRHGogkyu7kMYjA

Language: English

Subject area: Biotechnology

Doctoral student: Reskandi C. Rudjito , Glykovetenskap

Opponent: Associate Professor Mirjam Kabel, Wageningen University and Research

Supervisor: Universitetslektor Franscisco Vilaplana, Glykovetenskap; Doktor Amparo Jimenez-Quero, Glykovetenskap

QC 2022-02-09

Abstract

Cereal by-products are abundant streams of underutilised biomass that have the potential be valorised into various industrial products by means of environmentally friendly processes. In this study, an integrated bioprocess combining subcritical water extraction (SWE) and enzymatic treatment (ET) was proposed to effectively extract arabinoxylans (AX) from cereal by-products. Due to their unique substituted and feruloylated nature, AXs can potentially be developed into several functional food and materials, including packaging, coating and medical applications.

Optimisation of SWE was investigated in terms of how different extraction parameters could affect the extractability and the molecular structure of the AXs. The parameters investigated, included pretreatment, extraction time, biomass type, pH and temperature at laboratory scale, as well as upscaling at pilot scale (scale factor of 33x). The need for destarching prior to SWE was clearly highlighted as it improved not only the yield but also the purity of the extracted AX. Generally, an increase in temperature resulted in higher yields and a quicker transition from a glucan-rich to an AX-rich extract, but with the consequence of increased autohydrolysis. With increased time, AX populations generally increased in purity and complexity, exhibiting higher arabinose to xylose (A/X) ratios, but also lower molar mass distributions. By using a buffered solvent in neutral or mild alkaline conditions, autohydrolysis could be prevented. The choice of alkaline conditions was especially favourable in the release of glucuronoarabinoxylan (GAX) from corn fibre, but also significantly reduced the amount of esterified ferulic acid (FA) on the polysaccharide. The FA is an important moiety to preserve as it influenced the antioxidant activity of the extracts, as well as the possibility for oxidative coupling in material applications. In terms of biomass type, barley bran was found to be more recalcitrant than wheat or rye bran, displaying AX populations that were more heavily substituted and higher presence of p-coumaric acids in addition to FA. Ultimately, upscaling of the SWE revealed that similar results could be obtained to the laboratory scale when both the extraction time and solid to liquid ratios were increased. 

For the characterisation of enzymes, specifically xylanolytic enzymes, activity was first investigated on soluble AX substrates prior to the insoluble bran. Here, the effect of both arabinose substitution and feruloylation on AX substrates on the enzyme activity were investigated on less studied xylanases from GH family 5 and 8, with those that are well studied from GH family 10 and 11. The GH8 xylanase was most restricted by arabinose substitution, generating mid-length linear xylooligosaccharides (XOS) of up to a xylopentoase (X1-X5). Interestingly, the GH5s preferred highly substituted AXs, which was later attributed to the need of a C(O)-3 linked arabinose for cleavage. The GH5s distinctly produced arabinoxylan-oligosaccharides that were substituted on the reducing end. Meanwhile ferulic acid (FA), which occurs on C(O)-3 linked arabinoses, impeded the activity of the GH5 xylanases. In terms of efficiency, the traditional GH10 and GH11 xylanases were largely more active on a wider range of AX substrates, generating small hydrolysis products (X1-X3) at a much faster rate than the GH8 and GH5.

The knowledge obtained from the SWE optimisation, and the characterisation of enzymes were used to validate an integrated process combining the two approaches. The use of a GH10 xylanase together with a feruloyl esterase was later found as the most optimal cocktail in aiding the release of arabinoxylans in combination with SWE. It is anticipated that this study can enrich the knowledge on SWE technology, interactions between xylanolytic enzymes and arabinoxylan substrates, as well as the realisation of cereal-based biorefineries in the future.

urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-307789