Assembly of Functional Cellulolytic Enzymes

Assembly of Functional Cellulolytic Enzymes In the present study, we reported the assembly of functional cellulolytic enzymes using a synthetic, cell-surface engineered yeast consortium. Trichoderma reesei endoglucanase II (EGII) and cellobiohydrolase II (CBHII) and Aspergillus aculeatus ÃŽ ²-glucosidase I (BGLI) were displayed as fusion proteins with the AGA2p C-terminus of a-agglutinin on the cell surface of the diploid yeast strain Saccharomyces. cerevisiae Y5. The immobilization of each enzyme on the cell surface was confirmed by immunofluorescence microscopy. This type of yeast consortium allowed convenient optimization of ethanol production by adjusting the combination ratios of each cell type for inducing synergy in cellulose hydrolysis. Next, the direct ethanol fermentation from steam-exploded corn stover was investigated. The optimized cellulase-displaying consortium produced 20.4 g/l ethanol from 48.4 g cellulose per liter after 72 h in the presence of a small amount of cellulase reagent (0.9 FPU/ml). These findings suggested the feasibility of the cellulase-displaying yeast consortium for simultaneous saccharification and fermentation. Currently, many technological barriers exist with respect to the economical production of ethanol from lignocellulosic biomasses [1]. In the process of hydrolyzing cellulose into soluble sugars, multiple cellulases including endoglucanase (EG), cellobiohydrolase (CBH), and ÃŽ ²-glucosidase (BGL) are required [2]. Consolidated bioprocessing (CBP), which combines enzyme production, hydrolysis, and fermentation in one step, is a promising strategy for effective ethanol production from lignocellulosic materials. Saccharomyces cerevisiae is the traditional microorganism used for ethanol production, but it is unable to utilize cellulosic materials and a saccharification process is required prior to fermentation to produce glucose [3-4]. Numerous attempts have been made to engineer S. cerevisiae strains to express cellulases by cell surface engineering for direct ethanol production from cellulose, and although various bifunctional or trifunctional cellulose-degrading strains have been const ructed, the efficiency of cellulose degradation has not been sufficiently improved [5-9]. It would appear that co-expression of all cellulolytic enzymes in a single cell resulted in relatively low expression levels of cellulases, which may have been due to the heavy metabolic burden and potential jamming of the secretion machinery [6,7,10]. Therefore, in this study, we adapted a new strategy of performing simultaneous saccharification and fermentation with a synthetically engineered yeast consortium having the desired properties of cellulolytic ability and ethanol production to reduce the metabolic burden. The development of a diploid yeast strain is another promising strategy for improving expression levels of heterologous genes and enhancing the fermentation performance of S. cerevisiae. Because diploid strains have better growth ability as well as stress tolerances compared with haploid strains, they are particularly suited for industrial applications. Previously, our group reported on the construction of an à Ã‚ °-agglutinin expression system for genetic immobilization ÃŽ ²-glucosidase I on the cell surface of S. cerevisiae Y5 (Patent No: ZL200810222897.7, CGMCC2660). This diploid robust yeast strain possessed many advantages, such as higher ethanol yield, higher resistance to ethanol, and higher physiological tolerance to inhibitors present in lignocellulosic hydrolysates. Here, we report on our efforts to demonstrate the assembly of functional cellulolytic enzymes using a synthetic yeast consortium. In this study, we demonstrated the feasibility of constructing a novel cell surface engineered diploid yeast consortium for direct ethanol production from phosphoric acid swollen cellulose (PASC) and steam-exploded corn stover (CS), an important step toward direct ethanol production from insoluble cellulosic materials. The strains and plasmids used in this study are summarized in Table S1. Saccharomyces cerevisiae Y5 used for the yeast cell surface display of the cellulolytic enzymes was a newly developed diploid strain in our laboratory. E. coli Top 10 was used as the host strain for recombinant DNA manipulation. T. reesei was purchased from CICC (China Center of Industrial Culture Collection). E. coli transformants were grown in Luria-Bertani medium (1% tryptone, 0.5% yeast extract and 1% NaCl, pH 7.0) supplemented with 100 ug/ml of ampicillin. S. cerevisiae Y5 transformants were selected and maintained on Geneticin plates (1% yeast extract, 2% peptone and 2% glucose supplemented with 600 ug/ml Geneticin) at 30 °C , were induced in YPG (1% yeast extract, 2% peptone, and 2% galactose) at 20 °C. The fermentation medium was composed of 10 g/l yeast extract, 20 g/l polypeptone and 10 g/l PASC as the sole carbon source. The à ¯Ã‚ ¬Ã‚ lamentous fungus T. reesei was cultured in potato dextrose aga r medium (2% potato extract, 2% glucose) at 27 °C. The cDNA was synthesized from mRNA by using the First-Strand cDNA synthesis kit (Fermentas). Unless otherwise indicated, all chemicals, media components and supplements were of analytical grade standard and obtained from Sigma-Aldrich (St. Louis, MO, USA). All restriction enzymes were purchased from New England BioLabs (Ltd. Beijing). Primers used for plasmid construction are provided in Table S2. Plasmid pAGA1 for over-expression of the AGA1 gene and plasmid pBGLI for cell surface display BGLI were constructed previously [11]. Plasmid pEGII for cell surface expression of the EGII (egl2) was constructed as follows. The 1194 bp DNA fragment encoding the egl2 gene without its native secretion signal was amplià ¯Ã‚ ¬Ã‚ ed with the à ¯Ã‚ ¬Ã‚ rst-strand cDNA prepared from T. reesei as the template using primer pairs egl2-For/Rev, this DNA fragment was introduced into the yeast display vector pYD1(Invitrogen) with Kpn I/BamH I. MAT terminator was amplified from pYD1 by using primer pairs MAT-For/Rev and then digested with BamH I/EcoR I to create plasmid pYD1-egl2MAT. The KanR fragment was obtained from plasmid YIP5-KanR by two-step cloning. First, the DNA fragment containing ADH promoter and KanR ORF was amplified from YIP5-KanR by PCR using the KanR-For/Rev primers and inserted into EcoR I/Apa I site of plasmid pYD1-egl2MAT; next, the ADH terminator digested with Bgl II/Nde I was also introduced into pYD1-egl2MAT. The resulting plasmid was named pEGII. For displaying the T. reesei CBHII gene (cbh2) in S. cer evisiae Y5, plasmid pCBHII was created. A 1344 bp gene fragment coding for the mature region of the CBHII was amplià ¯Ã‚ ¬Ã‚ ed using primers cbh2-For/Rev-KT and introduced into plasmid pEGII digested with Kpn I/BamH I for replacing egl2 to form pCBHII (Figure 1). Transformation of S. cerevisiae Y5 was carried out using the lithium acetate method [12]. The plasmid pAGA1 was linearized by Apa I for chromosome integration. The plasmid pYD1 was transformed into S. cerevisiae Y5 as a negative control. S. cerevisiae Y5 clones transformed with different plasmids (strain Y5/pYD1 contained plasmids pAGA1 and pYD1, strain Y5/EGII contained plasmids pAGA1 and pEGII, strain Y5/CBHII contained plasmids pAGA1 and pCBHII) were selected and maintained on Geneticin(G418) plates. Immunofluorescence microscopy was performed as described previously [13]. Immunostaining was performed as follows. Induced recombinant yeast cells expressing cellulases were harvested by centrifugation at 6000 rpm for 5 min and washed with phosphate-buffered saline (PBS). As the primary antibody, mouse anti-Xpress tag antibody (Invitrogen, R910-25) for EGII and CBHII was used at dilution rates of 1:1000. As the second antibody, Fuorescein (FITC)-conjugated goat anti-mouse IgG(H+L) (Jackson, 115-095-003) was used at dilution rate 1:200. Cells and the anti-body were incubated at room temperature. After washing the cell–antibody complex with PBS twice, cellular localizations of the cellulases were observed under a fluorescence microscope. Yeast strains Y5 and Y5/pYD1were used as control. Yeast cells were induced in YPG medium for 48 h at 20 ºC and harvested by centrifugation for 5 min at 6000 rpm, washed with distilled water. BGLI activity of strain Y5/BGLI was measured using à ¯Ã‚ Ã‚ ²-nitrophenyl-ÃŽ ²-D-glucopyranoside as the substrate according to a previously described method [14]. Endoglucanase and cellobiohydrolase activities were determined by hydrolysis of carboxymethyl cellulose (CMC) and phosphoric acid swollen cellulose (PASC), respectively. PASC was prepared from Avicel PH-101 (Fluka Chemie GmbH, Buchs, Switzerland) as amorphous cellulose. The cell pellet was resuspended in a reaction mixture of 1% CMC or 1% PASC in 50 mM sodium acetate buffer (pH 5.0) with the optical density at 600 nm adjusted to 1.0. After a reaction at 50 ºC for 30 min, the activities were determined by DNS method [15]. One unit of enzyme activity was defined as the amount of enzyme released 1 ÃŽ ¼mol reducing sugar from the substrate per minute. The abilities of the engineered yeast consortium (Y5/EGII + Y5/CBHII + Y5/BGLI) to fermentation ethanol from PASC and steam-exploded corn stover were investigated. The steam-exploded corn stover used in this study was provided by Henan Tian Guan Group Co., Ltd (Nanyang, Henan, China). The raw material was chopped to 2-3 cm size and treated in a steam-exploded vessel at 2.0 MPa for 5 min. The pretreated feedstock was dried at room temperature and directly used as a substrate without washing. The moisture content of the substrate was 8%. The composition of materials was quantitatively analyzed following the NREL Laboratory Analytical Procedure NREL/TP-510-42618 (Structural carbohydrates and lignin) (Sluiter et al., 2008)[16], as shown in Table 3. An enzyme mixture composed of equal amounts of cellulase (Sigma-Aldrich, St. Louis, MO) and ÃŽ ²-glucosidase (Sigma-Aldrich) was used. Yeast cells harboring different surface-display plasmid for EGII, CBHII, or BGLI, were grown in YPD medium a nd then transferred to YPG medium for 48 h at 20 ºC to express cellulase. Cells collected by centrifugation at 5000 rpm for 5 min at 4 ºC, washed with distilled water twice, and mixed in the adjustable ratio to a total initial cell concentration of 30 g/l wet weight to form the functional consortium. Ethanol fermentation proceeded at 30 ºC with 90 rpm in 250 ml Erlenmeyer flasks. 1ml samples of the fermentation broth were taken periodically and stored at -4 ºC until they were analyzed for sugar and ethanol content. The total sugar was determined by the phenol-sulfuric acid method [17]. Glucose was measured by HPLC (model 1260, Agilent Technologies) equipped with a Hi-Plex H column 300 mm Ãâ€" 7.7 mm) and a refractory index (RI) detector. Samples were run at a temperature of 60 ºC and a mobile phase of 5 mM sulfuric acid at a flow rate of 0.6 ml/min. Ethanol analysis was carried out using GC (model 7890A, Agilent Technologies) equipped with a flame ionization detector and a HJ-PEG column. Samples were run under the following conditions: column oven at 120 ºC, front injection port at 200 ºC, with N2 as the carrier gas at a flow rate of 4 ml/min. The expression plasmids pEGII and pCBHII (Fig. 1) were transformed into the yeast S.cerevisiae Y5 strains, respectively. All of recombinant yeast strains had a pAGA1 plasmid for integrating AGA1 into the chromosome, and the resultant transformants were designated strains Y5/EGII and Y5/CBHII (Table S1). Upon galactose induction, the proteins were expected to be secreted and interact with the Aga1p and Aga2p anchor system by using the glycosylphosphatidylinositol (GPI) anchor linked to the cell surface. To confirm displaying of EGII and CBHII on the yeast cell surface, immunofluorescence labeling of the cells was carried out using mouse anti-Xpress IgG antibody as the primary antibody. The green fluorescence of Fuorescein (FITC)-conjugated goat anti-mouse IgG was observed for strains Y5/EGII and Y5/CBHII (Fig. 2), indicating that EGII and CBHII were displayed on the cell surface, respectively. The cells harboring the control plasmids were hardly labeled with mouse anti-Xpress IgG(Fig. 2). These results suggested that two types of cellulase were successfully expressed on the cell surface of S. cerevisiae Y5 strain. As shown in Table 1, EGII, CBHII and BGLI activities were detected in the pellet fraction of strain Y5/EGII, Y5/CBHII and Y5/BGLI, respectively. The strain Y5/CBHII and strain Y5/EGII showed moderate CBHII and EGII activity (1.14 U/OD600 and 1.27 U/OD600, respectively). The BGLI activity of strain Y5/BGLI cells was relatively low, which was only 0.72 U/OD600. No enzyme activity was detected in the culture supernatant (data not shown), and the control strain without displayed enzymes exhibited less than 0.1 U/OD600 of enzyme activity. These results clearly indicated that active enzymes were displayed on the cell surface without leakage into the culture medium. Ethanol fermentation from 10 g amorphous cellulose per liter was performed using a cell combination system consisted of three cellulase-displaying yeast populations. Cells displaying EGII, CBHII and BGLI were mixed in various ratios and the produced ethanol from PASC were measured. S.cerevisiae Y5 without displayed enzymes was the control strain. A mixture of cells with EGII: CBHII: BGLI ratio of 2:1:1 produced the highest amount of ethanol (1.76 g/l) after 84 h; the yield (in grams of ethanol produced per gram of consumed reducing sugar) was 0.42 g/g (Fig. 3). A mixture of cells composed of an equal amount of each cell type produced 0.68 g/l ethanol after 84 h (Figure 3), indicating about 1.6-fold improvement of ethanol production by optimizing the cell ratio. However, a large portion of the substrate (the amount of residual sugar after 84 h hydrolysis of 10 g/l PASC was 5.5 g/l, and the sugar consumption rate was 43.3%) remained after 96 h without being hydrolyzed because the cellu lase activities displaying on cell surface were not enough for complete cellulose digestion. Simultaneous saccharification and fermentation of steam-exploded corn stover (CS) as a sole carbon source was conducted for the cellulase-displaying yeast consortium of the optimized ratio 2:1:1 in the presence of commercial cellulase (Sigma-Aldrich, St. Louis, MO) with different enzyme loadings (0, 0.3, 0.6, 0.9, 1.2, 1.5 FPU/ml). A mixture of cells was incubated in 100 ml of YP medium (20 g/l peptone, 10 g/l yeast extract) for 1 h to remove residual carbon source, and then resuspended in YP-CS medium (YP medium containing 100 g/l steam-exploded corn stover, corresponding to 48.4g cellulose per liter). As shown in Fig. 4, in the presence of 0, 0.3, 0.6, 0.9, 1.2 and 1.5 FPU/ml cellulase, 34.49, 18.71, 7.03, 2.11, 1.98, and 1.23 g/l of residual cellulose remained after 84h, respectively. Addition of 0.9 FPU/ml cellulase enabled utilization of 92.3% of the initial cellulose (Figure 4). The cellulose hydrolyzed by cellulase-displaying yeast consortium with an additional 0.9 FPU/ml cellulase was nearly the same as that by control strain S.cerevisiae Y5 with an additional 1.5 FPU/ml. These results indicate that cellulases displayed on the yeast cell surface improve hydrolysis of cellulose, although their activities were lower than commercial enzymes. Furthermore, using the optimized cell combination system, the relationship between the amount of added cellulase and final ethanol concentration was investigated. As shown in Fig. 5, in the presence of 0.9 FPU/ml cellulase, the cellulase-displaying consortium produced 20.4 g/l ethanol after 72 h, which was similar to the value (20.9 g/l) obtained by control strain in the presence of 1.5 FPU/ml cellulase (Table 2). Notably, as the ethanol yield reached 86% of the theoretical yield with 0.9 FPU/ml cellulase, the cell-surface engineered system enabled a reduction in the amount of added commercial cellulase. Hydrolysis of crystalline cellulose to glucose requires the sequential reactions of three groups of cellulases: endoglucanase, cellobiohydrolase, and ÃŽ ²-glucosidase. CBP is a one-step process where all steps occur in a single reactor and a single microorganism or microbial consortium converts pretreated biomass to ethanol with no additional commercial enzymes. The key challenge of CBP lies in choosing the optimal host to directly convert lignocellulosic materials to ethanol. In recent years, several researchers have been engaged in co-displaying multiple cellulases in a single cell for direct conversion of cellulose to ethanol [18-21]. However, the enzyme activity can be limited because of the metabolic burden [22]. Furthermore, it is difficult to control the surface expression level of each enzyme for optimal ethanol fermentation. Apiwatanapiwat et al., constructed the engineered yeast strain NBRC-5Es that co-displayed two types of amylolytic enzymes, two types of cellulolytic enz ymes (T. reesei EGII and CBHII), and A. aculeatus BGLI on the cell surface. The NBRC-5Es strain produced 1.04 g/l ethanol from 8.44 g/l of the acid-treated Avicel after 48 h of fermentation and resulted in a large portion of the substrate remaining without being hydrolyzed by the enzymes. In this study, instead of co-displaying all the enzymes in one cell, we developed a cellulase-displaying yeast consortium consisting of three types of yeast cells, each displaying different cellulases. This method allows for convenient optimization of ethanol production by adjusting the combination ratio of each cell type for inducing a synergy in cellulose hydrolysis. Diploidization is also a promising strategy for enhancing the fermentation ability of S. cerevisiae. Because polyploid yeast strains, including diploid strains, have higher cell growth rates, cell yields, and tolerances to various stresses compared with haploid strains, they are particularly suited for industrial application. Therefore, to generate an efficient â€Å"whole-cell biocatalyst† yeast strain related to cellulosic ethanol production, we selected S. cerevisiae Y5, a robust diploid strain, as the host cell based on its fermentation and inhibitor tolerance properties [23-24]. We first explored the possibility of ethanol fermentation from PASC by using the surface-immobilized yeast consortium (Y5/EGII+Y5/CBHII+Y5/BGLI). A mixture of cells at the optimized EGII: CBHII: BGLI ratio of 2:1:1 produced 1.6-fold more ethanol (1.76 g/l) than cells composed of an equal amount of each cell type. Next, the fermentation performance of yeast consortium using steam-exploded CS as the sole carbon source was further investigated. The optimized cellulase-displaying consortium produced 20.4 g/l ethanol from 48.4 g cellulose per liter after 72 h in the presence of a small amount of cellulase reagent (0.9 FPU/ml), suggesting the feasibility of the cellulase-displaying yeast consortium for simultaneous saccharification and fermentation. Although several studies have been carried out on establishing a cell-displaying yeast consortium [25-27], few reports of direct ethanol fermentation from pretreated lignocellulosic material have been published. The combined cell system describ ed here could become the basis for the eventual direct ethanol production from insoluble cellulosic materials.