Practical ethanol production from starchy biomass such as cassava, rice, sweet sorghum, and sweet potato have been reported; of these, corn is the most commonly used starchy feedstock for bioethanol production. This process uses starch from grains resembling corn; it is gelatinized by cooking and hydrolyzed to form glucose, which can be fermented by the microorganisms. Conventional ethanol production from raw starch requires the following three steps: (1) cooking at high temperature (140–180◦C), (2) addition ofα-amylase and glucoamylase for enzymatic saccharifi- cation of the cooked materials to glucose, and (3) fermentation of glucose to ethanol.
The liquefaction process, which accounts for 30–40% of the total energy used for ethanol production, combined with the large quantities of enzymes that are required to convert the raw starch into glucose, both contribute to making conventional ethanol production an expensive and complex process. Matsumoto et al. (1982) reported non-cooking and low temperature cooking fermentation systems to reduce energy consumption by approximately 50%, though it is still necessary to add large amounts of amylolytic enzymes to hydrolyze the starchy materials to glucose.
Many researchers have attempted to resolve the problem by using recombinant amylase-expressing yeasts with the ability to directly ferment starch to ethanol (Tamalampudi et al., 2009). These amylases can be generally defined as the enzymes that hydrolyze theO-glycosyl linkage of starch (Nair et al., 2009).α-Amylases that cleave the internalα-1,4-glucosidic linkages in the starch are one of the most pop- ular and important forms of industrial amylases, which are chiefly required for the thinning of starch in the liquefaction process. Glucoamylases are capable of cleav- ing both α-1,6- and α-1,4-glucoside linkages in the low molecular maltodextrins, removing one glucose unit at a time from the nonreducing end of the carbohydrate molecule. Recombinant yeast strains that coproduceα-amylase and glucoamylase have been developed to improve the efficiency of starch fermentation (Eksteen et al., 2003). Recent advances in yeast cell–surface engineering have provided the tools for the display of amylolytic enzymes that allow the utilization of yeast as a whole- cell biocatalyst for direct ethanol production from starch (Tamalampudi et al., 2009;
Yamada et al., 2010a). Moreover, the integration of hydrolysis and fermentation steps by surface engineering of yeast cells can reduce the unit operation costs compared to separate hydrolysis and fermentation (SHF) process.
8.2.2 Yeast Cell–Surface Engineering System for Biomass Utilization Yeast cell–surface engineering has been established to display enzymes, functional proteins, antibodies, and combinatorial protein libraries (Kondo and Ueda, 2004).
The cell surface is a functional interface between the inside and outside of the cell allowing some surface proteins to extend across the plasma membrane, while others are bound by non-covalent or covalent interactions to the cell surface components.
For anchoring surface-specific proteins, yeast cells have molecular systems to confine
proteins to particular domains on the cell surface (Pittet and Conzelmann, 2007). For instance, agglutinin (Agα1 and Aga1), flocculin Flo1, Sed1, Cwp1, Cwp2, Tip1, and Tir1/Srp1 all have glycosylphosphatidylinositol (GPI) anchor moieties that are covalently attached to the carboxy termini of proteins (Lesage and Bussey, 2006).
The GPI-anchored proteins contain hydrophobic peptides at their carboxy termini.
After the completion of protein synthesis, the precursor protein remains anchored in the endoplasmic reticulum (ER) membrane by the hydrophobic carboxy-terminal sequence, with the rest of the protein in the ER lumen. Within less than a minute, the hydrophobic carboxy-terminal sequence is cleaved at theωsite and concomitantly replaced with a GPI anchor, presumably by the action of a transamidase (Ueda and Tanaka, 2000).
Biotechnology allows the cell surface to be exploited using yeast’s natural mech- anisms for anchoring proteins onto the cell surface (Figure 8.2). The use of the GPI-anchoring system has enabled display of various kinds of functional proteins on the cell surface without the loss of their activity through genetic engineering (Kondo and Ueda, 2004). Among the GPI anchor proteins,α-agglutinin and flocculin have been mainly used for the display of biomass-degrading enzymes. In theα-agglutinin system, the carboxy-terminal half of the α-agglutinin containing the GPI anchor attachment signal is connected to a target enzyme to anchor it as a fusion protein on the yeast cell surface. In the flocculin system, two types of cell surface display methods have been developed. In one system, the carboxy-terminal region of Flo1p, which contains a GPI attachment signal, is used, while the second system attempts to utilize the ability of the flocculation functional domain of Flo1p to create a novel surface display apparatus (Kondo and Ueda, 2004).
An advantage of the cell-surface engineering system is that biomass-degrading enzymes are genetically self-immobilized on the yeast cell surface so that the activities of the enzymes are retained as long as the yeast continues growing, whereas it is not easy to maintain the activities for a long reaction period in the conventional direct fermentation system in which the enzymes are secreted into the medium (Ueda and Tanaka, 2000). Another advantage of this system is the easy separation of the biocatalyst from the product. Reutilization of the yeast cells enables reuse of the enzymes displayed on the cell surface without reproduction of the yeast cells, which would reduce the cost of yeast propagation as well as enzyme addition (Kondo et al., 2002; Matano et al., 2013a). Saccharomyces cerevisiae is useful as a host for genetic engineering, since it allows the folding and glycosylation of expressed heterologous eukaryotic proteins and can be genetically manipulated. Moreover, the yeast can be cultivated to a high density in an inexpensive medium, so that the display of enzymes on the cell surface can have several applications in bioconversion processes.
8.2.3 Ethanol Production from Starchy Biomass Using Amylase-Expressing Yeast
Surface expression of amylolytic enzymes was initiated by the pioneering work of Murai et al. (1997) who reported a strategy for development of recombinant Sa.
cerevisiae strains displaying amylolytic enzymes on the cell surface. Glucoamy- lase derived fromRhizopus oryzaeglucoamylase was displayed as a fusion protein withα-agglutinin using the secretion signal peptide ofRh. oryzaeglucoamylase on the cell surface of a laboratorySa. cerevisiaestrain, MT8-1. The anchoring on the cell wall was verified by immunofluorescence labeling with anti-glucoamylase IgG.
Kondo et al. (2002) used a flocculating yeast strain, YF207, for the surface display of glucoamylase because such a yeast strain offers the advantage of easy harvest of cells with displayed enzymes by flocculation after the batch fermentation. In this experiment, the display of glucoamylase did not negatively affect the growth and flocculation ability of the yeast during the ethanol production phase. Moreover, the activity of the glucoamylase displayed on the surface of flocculent yeast was similar to that displayed on non-flocculent yeast. The recombinant glucoamylase-displaying flocculent yeast strain maintained a high ethanol production rate of 0.6–0.7 g/L⋅h during repeated batch fermentation of soluble starch over 300 hours. In the fermen- tation by glucoamylase-displaying yeast cells, glucose was maintained at a very low concentration, which might be because the recombinant yeast cells metabolize glu- cose as soon as it is released from soluble starch by the glucoamylase. This low concentration of glucose in the fermentation is advantageous in minimizing the risk of contamination. On the other hand, the display of only glucoamylase led to the accumulation of an insoluble starch fraction during fed-batch fermentation because of the lack of a liquefying enzyme,α-amylase. In order to overcome this problem, Shigechi et al. (2002) developed two recombinant yeast strains: one co-displaying glucoamylase andα-amylase on the cell surface and the other displaying glucoamy- lase and secretingα-amylase into the culture medium. In fed-batch fermentations with soluble potato starch as a substrate, these two recombinant strains produced more than 60 g/L ethanol after 100 hours under anaerobic conditions. However, in using theα-amylase-secreting strain, glucose concentration in the culture medium was slightly higher, which was probably due to higher accessibility of starch to the secretedα-amylase in the medium.
Ethanol production from low temperature cooked starch has several advantages over the conventional high temperature cooking process because the high tempera- ture cooking requires high energy and the addition of large amounts of amylolytic enzymes. Shigechi et al. (2004a) directly produced ethanol from corn starch cooked at 80◦C in a single step using a recombinant yeast strain co-displaying both glucoamy- lase fromRh. oryzaeandα-amylase fromBacillus stearothermophilus.α-Amylase hydrolysis ofα-1,4-linkages of starch in a random fashion played an important role in the cooperative and sequential decomposition of the starch, which led to the effi- cient production of ethanol. The maximum ethanol titer, ethanol production rate, and substrate consumption rate from the fermentation of low-temperature cooked starch were almost the same as those from high-temperature cooked starch. Using the co-displaying strain, the yield of ethanol produced was 0.50 g per gram of car- bohydrate consumed, which corresponds to 97.2% of the theoretical yield (0.51 g of ethanol per gram of glucose). The isolation of amylase from the lactic acid bac- teriumStreptococcus bovisopened new horizons for the hydrolysis of raw starch.
Shigechi et al. (2004b) developed a novel non-cooking fermentation system for direct
ethanol production with a yeast strain co-displayingRh. oryzaeglucoamylase and St. bovisα-amylase usingα-agglutinin and Flo1p, respectively, as anchor proteins.
They reported thatα-amylase activity depends on the anchor protein. Glucoamylase activity is anchor independent while the use of the Flo1 anchor led to 40 times higher α-amylase activity than with theα-agglutinin anchor. Since severalα-amylases have a starch-binding domain at their carboxy-terminal region, the display ofα-amylase using Flo1p as anchor would be effective for ethanol production from raw corn starch.
The yield, in terms of grams, of ethanol produced per gram of sugar consumed was 0.44 g/g, which corresponds to 86.5% of the theoretical yield (Shigechi et al., 2004b).
Recently, Yamada et al. (2010a) constructed a high-performance, starch-degrading yeast capable of direct ethanol production from purified raw corn starch by combining theδ-integration technique and polyploidization with high yields. In addition, brown rice was directly converted to ethanol by the polyploid amylase-expressing yeast strain without any pretreatment or addition of enzymes or nutrients (Yamada et al., 2011b). Yamakawa et al. (2012) reported repeated batch fermentation of raw starch using a recombinantSa. cerevisiaestrain co-displayingα-amylase and glucoamylase.
Fermentation from 100 g/L of raw starch was repeated for 23 cycles without the loss of enzyme activity. By displaying amylolytic enzymes on yeast cell surfaces, efficient whole-cell biocatalysts for SSF have been constructed.