A wheat biorefining strategy based on solid-state fermentation for fermentative production of succinic acid Chenyu Du a, Sze Ki Carol Lin a, Apostolis Koutinas a, Ruohang Wang a, Pilar Dorado a,b, Colin Webb a,* a Satake Centre for Grain Process Engineering, School of Chemical Engineering and Analytical Science, University of Manchester, P.O. Box 88, Manchester M60 1QD, United Kingdom b Department of Physical Chemistry and Applied Thermodynamics, EPS, University of Cordoba, Spain a r t i c l e i n f o Article history: Received 15 November 2007 Received in revised form 28 February 2008 Accepted 1 March 2008 Available online 22 April 2008 Keywords: Succinic acid Solid-state fermentation (SSF) Wheat-based biorefinery Actinobacillus succinogenes Fungal fermentations a b s t r a c t In this study, a novel generic feedstock production strategy based on solid-state fermentation (SSF) has been developed and applied to the fermentative production of succinic acid. Wheat was fractionated into bran, gluten and gluten-free flour by milling and gluten extraction processes. The bran, which would nor- mally be a waste product of the wheat milling industry, was used to produce glucoamylase and protease enzymes via SSF using Aspergillus awamori and Aspergillus oryzae, respectively. The resulting solutions were separately utilised for the hydrolysis of gluten-free flour and gluten to generate a glucose-rich stream of over 140 g l 1 glucose and a nitrogen-rich stream of more than 3.5 g l 1 free amino nitrogen. A microbial feedstock consisting of these two streams contained all the essential nutrients required for succinic acid fermentations using Actinobacillus succinogenes. In a fermentation using only the combined hydrolysate streams, around 22 g l 1 succinic acid was produced. The addition of MgCO3 into the wheat- derived medium improved the succinic acid production further to more than 64 g l 1. These results dem- onstrate the SSF-based strategy is a successful approach for the production of a generic feedstock from wheat, and that this feedstock can be efficiently utilised for succinic acid production. �� 2008 Elsevier Ltd. All rights reserved. 1. Introduction Current technologies for the production of organic chemicals, fuels and polymers depend mainly on fossilized hydrocarbons. However, the growing demand for energy, the depletion of fossil resources and the increasing concern of environmental issues have created the need for the development of sustainable alternatives based on renewable raw materials (Webb et al., 2004). The selec- tion of appropriate raw materials and the development of biorefin- ery-based strategies to support sustainable processes are therefore essential. In the UK, cereals (and wheat in particular) provide po- tential as renewable raw materials (Koutinas et al., 2004). Although wheat contains all the nutrients generally required for bacterial fermentations for the production of platform chemicals, these are not in a directly accessible form. Therefore, some pre- processing is required in order to release these nutrients. The approaches to exploit wheat components for the produc- tion of platform chemicals can be generally categorized as follows: (i) processing wheat into relatively pure starch which is then lique- fied and saccharified to liberate glucose to produce fine chemicals (Hofvendahl et al., 1999 Hussy et al., 2003) (ii) directly utilising wheat flour using certain kinds of microorganisms (Zhang and Cheryan, 1991 Okano et al., 2007) (iii) exploring a biorefining strategy to convert all the nutrients of wheat into a generic feed- stock and then to ferment the generic feedstock for the production of fine chemicals (Webb et al., 2004 Koutinas et al., 2004, 2007b). However, the first approach only utilises starch without exploring uses for the other components in wheat. It also involves the use of expensive commercial enzymes, which implies high raw material cost. The second approach is limited to microorganisms that can directly degrade starch and thus only few examples are available, to date, the most representative of which are in lactic acid and bio-ethanol fermentations (Zhang and Cheryan, 1991 Khaw et al., 2006 Okano et al., 2007). For the third approach, a wheat biorefining strategy has been developed based on fungal fermenta- tion (Webb et al., 2004), and this submerged fermentation (SmF) strategy has been applied to the production of biofuels, biodegrad- able polymers and platform chemicals (Du et al., 2007 Koutinas et al., 2007a,c). In the wheat-based biorefining strategy mentioned above, complete starch hydrolysis has been demonstrated but, on the other hand, protein hydrolysis is not efficient (Koutinas et al., 2007a). Complete starch and protein hydrolysis would create the potential to formulate media for a spectrum of fermentations (Kou- tinas et al., 2007b). However, complete hydrolysis of gluten can only be achieved by employing a microorganism highly efficient in the production of proteolytic enzymes, such as Aspergillus ory- zae. Using two submerged fermentations to produce amylolytic and proteolytic enzymes would lead to higher amounts of wheat 0960-8524/$ - see front matter �� 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.biortech.2008.03.019 * Corresponding author. Tel.: +44 (0) 161 306 4379 fax: +44 (0) 161 306 4399. E-mail address: colin.webb@manchester.ac.uk (C. Webb). Bioresource Technology 99 (2008) 8310���8315 Contents lists available at ScienceDirect Bioresource Technology journal homepage: www.elsevier.com/locate/biortech

flour consumption for growing these microorganisms, and to high- er operating cost. In comparison to SmF, solid-state fermentation (SSF) could utilise efficiently many cheap raw materials including wheat bran and possibly improve process economics (Blandino et al., 2002 Pandey et al., 1999). This paper presents a novel wheat-based bior- efining strategy based on the SSF of wheat bran (Fig. 1). The two SSF carried out by Aspergillus awamori and A. oryzae were integrated with flour and gluten hydrolysis, respectively, for the production of generic feedstock media. Mixtures of flour hydrolysate and glu- ten hydrolysate were then examined for fermentative production of succinic acid. Any remaining gluten hydrolysate could be used in fermentations of recombinant Escherichia coli cultures to produce recombinant peptides or proteins (unpublished data). Succinic acid is an important 4-C building block, which is widely recognized as a potential platform chemical for the produc- tion of various value-added derivatives (McKinlay et al., 2007). However, compared with the traditional chemical production pro- cesses, the fermentative production of succinic acid is still expen- sive. One possible solution is to exploit the feasibility to produce succinic acid in a biorefining strategy utilising renewable raw materials so as to replace the utilisation of expensive commercial glucose. Therefore, the substitution of synthetic media by the gen- eric feedstock, produced from the SSF-based strategy, presents the possibility of decreasing production costs and creating an econom- ically viable bioprocess. 2. Methods 2.1. Microorganisms A. awamori and A. oryzae were utilised in SSF to produce amylo- lytic and proteolytic enzymes, respectively. Their storage and spor- ulation for inoculum preparation have been described in previous publications (Koutinas et al., 2001 Wang et al., 2005). Actinobacil- lus succinogenes ATCC 55618 was utilised for succinic acid fermen- tations (Du et al., 2007). 2.2. Raw materials The wheat used in this study was a soft variety, Consort. The composition of the wheat has been described in a previous publi- cation (Koutinas et al., 2007a). The wheat was processed through a B��hler laboratory mill to produce wheat flour and bran. To ex- tract gluten, the conventional Martin process was applied (MacRit- chie, 1985). The amount of water used in this process was about 1 l per 200 g wheat flour (dry basis). The remaining suspension was designated as Gluten-Free Flour Suspension (GFFS). 2.3. Solid-state fermentation by A. awamori and A. oryzae Two Duran bottles each containing 90 g wheat bran were auto- claved at 121 ��C for 30 min. After the moisture content of the bran was brought up to 60% by the addition of sterile water, the bottles were separately inoculated with A. awamori and A. oryzae to achieve the target inoculum sizes (4 106 and 1 106 spores per g wheat bran for A. awamori and A. oryzae, respectively). Approxi- mately 15 g of wheat bran was then distributed into each of six 14 cm Petri dishes and incubated at 30 ��C for 96 h in the case of A. awamori or 48 h in the case of A. oryzae. 2.4. Enzyme extraction Fermented mashes in each Petri dish were mixed with distilled water at a ratio of 1 g fermented mash to 10 ml distilled water. En- zyme extraction was carried out in a 1-l Duran bottle at room tem- perature for 1 h and mixed by a magnetic stirrer. Filtrate through Whatman No. 1 filter paper was collected as enzyme solution, and was applied in the following flour and gluten hydrolysis. 2.5. Flour and gluten hydrolysis The glucoamylase-rich solution was used to hydrolyse GFFS to produce a glucose-rich stream (designated as Gluten-Free Flour Flour hydrolysis Gluten hydrolysis SSF by A. awamori Flour Hydrolysate Gluten hydrolysate Bacterial Fermentation Wheat bran SSF by A. oryzae Gluten-free flour Gluten Glucoamylase solution Protease solution Succinic acid Bacterial Fermentation Recombinant proteins Milling and gluten extraction Wheat Fig. 1. Schematic diagram of the SSF-based biorefining strategy for converting wheat to succinic acid. Wheat is milled into flour and bran. The flour is separated into a GFFS (consisting mainly of starch) and gluten. Mixtures of flour and gluten hydrolysates are used in succinic acid fermentations. C. Du et al. / Bioresource Technology 99 (2008) 8310���8315 8311