TXT

Ethanol fermentation from lignocellulosic hydrolysate

By Marie Greene,2014-08-25 18:39
8 views 0
Ethanol fermentation from lignocellulosic hydrolysate

     ??ÎÄÓÉhailong870318??Ï×

    pdfÎĵµ?ÉÄÜÔÚWAP?Ëä?ÀÀÌåÑé???Ñ????ÒéÄúÓÅÏÈÑ?ÔñTXT???òÏÂÔØÔ?ÎÄ?þµ????ú?é????

     Appl Microbiol Biotechnol (2006) 72: 1136?C1143 DOI

    10.1007/s00253-006-0402-x

     BIOTECHNOLOG ICA L PROD UCTS A ND PRO CESS ENGINE ERIN G

     Satoshi Katahira . Atsuko Mizuike . Hideki Fukuda . Akihiko Kondo

     Ethanol fermentation from lignocellulosic hydrolysate by a recombinant xylose- and cellooligosaccharide-assimilating yeast strain

     Received: 8 December 2005 / Revised: 26 February 2006 / Accepted: 5 March 2006 / Published online: 31 March 2006 # Springer-Verlag 2006

     Abstract The sulfuric acid hydrolysate of lignocellulosic biomass, such as wood chips, from the forest industry is an important material for fuel bioethanol production. In this study, we constructed a recombinant yeast strain that can ferment xylose and

    cellooligosaccharides by integrating genes for the intercellular expressions of xylose reductase and xylitol dehydrogenase from Pichia stipitis, and xylulokinase from Saccharomyces cerevisiae and a gene for displaying ?Â-glucosidase from Aspergillus acleatus on the cell surface. In the fermentation of the sulfuric acid hydrolysate of wood chips, xylose and cellooligosaccharides were completely fermented after 36 h by the recombinant strain, and then about 30 g/l ethanol was produced from 73 g/l total sugar added at the beginning. In this case, the ethanol yield of this recombinant yeast was much higher than that of the control yeast. These results demonstrate that the fermentation of the lignocellulose hydrolysate is performed efficiently by the recombinant Saccharomyces strain with abilities for xylose assimilation and cellooligosaccharide degradation. Keywords Fermentation . Yeast . Xylose . Lignocellulosic hydrolysate . Ethanol . Cell surface display

     Introduction

     Renewable lignocellulosic biomass, such as agricultural and forestry residues, waste paper, and industrial waste, is an attractive feedstock for bioethanol production. The utilization of bioethanol as fuel can significantly suppress the accumulation of greenhouse gasses (McMillan 1997; Claassen et al. 1999). Lignocellulose, which is composed of cellulose, hemicellulose, and lignin (Aristidou and Penttil 2000; Saha 2003), is often hydrolyzed by acid treatment; the hydrolysate obtained is then used for ethanol fermentation by microorganisms such as yeast. Because such lignocellulose hydrolysate contains not only glucose, but also various monosaccharides, such as xylose, mannose, galactose, and arabinose, and oligosaccharides, microorganisms should

    be required to efficiently ferment these sugars for the successful industrial production of ethanol. One of the most effective ethanol-producing yeasts, Saccharomyces cerevisiae, has several advantages owing to its high ethanol production from hexoses and high tolerance to ethanol and other inhibitory compounds in the acid hydrolysates of lignocellulosic biomass (Olsson and Hahn-Hgerdal 1993; Hahn-Hgerdal et al. 2001). However, because wild-type strains of this yeast cannot utilize pentoses, such as xylose and arabinose, and celloligosaccharides, ethanol production from a lignocellulose hydrolysate is inadequate. Accordingly, many researchers have engineered yeast capable of xylose fermentation (Ho et al. 1998; Eliasson et al. 2000; Hahn-Hgerdal et al. 2001, Jeffries and Jin 2004). Natural xylose-fermenting yeasts, such as Pichia stipitis (Verduyn et al. 1985), Candida shehatae (Ho et al. 1990), and Candida parapsilosis (Lee et al. 2003), can metabolize xylose via the action of xylose reductase (XR) to convert xylose to xylitol, and of xylitol dehydrogenase (XDH) to convert xylitol to xylulose. Therefore, ethanol fermentation from xylose can be successfully performed by recombinant S. cerevisiae carrying heterologous XR and XDH from P. stipitis, and xylulokinase (XK) from S. cerevisiae (Ho et al. 1999; Eliasson et al. 2000; Toivari et al. 2001).

     S. Katahira . H. Fukuda Division of Molecular Science, Graduate School of Science and Technology, Kobe University, 1-1 Rokkodaicho, Nada-ku, Kobe 657-8501, Japan A. Mizuike . A. Kondo (*) Department of Chemical Science and Engineering, Faculty of Engineering, Kobe University, 1-1 Rokkodaicho, Nada-ku, Kobe 657-8501, Japan Tel.: +81-78-8036196 Fax: +81-78-8036196 e-mail: akondo@kobe-u.ac.jp

     1137

     On the other hand, for the fermentation of cellulosic materials to ethanol, various cellulases and ?Â-glucosidases have been expressed in S. cerevisiae (Murai et al. 1998; van Rensburg et al. 1998; Cho and Yoo 1999; Fujita et al. 2004). Also, the hydrolysis of

    cellooligosaccharides (two to six glucose units) and an efficient ethanol production from cellobiose by a recombinant yeast strain displaying Aspergillus aculeatus ?Â-glucosidase 1 (BGL1) on its cell surface have been achieved through cell-surface engineering (Murai et al. 1998; Fujita et al. 2002). Although the fermentations of xylose and cellobiose are simultaneously performed by the thermotolerant methylotrophic yeast Hansenula polymorpha (Ryabova et al. 2003), the ethanol yield and productivity are insufficient. In this study, we constructed a recombinant Saccharomyces strain with xylose-fermenting ability by introducing genes of P. stipitis XR (XYL1) and XDH (XYL2), and S. cerevisiae XK (XKS1) for their intracellular expression, and with cellooligosaccharide-degrading ability by introducing the fusion

    gene of A. aculeatus BGL1 and 3'-half of ?Áagglutinin for its cell-surface display. Using this recombinant strain, we attempted ethanol fermentation from a lignocellulose hydrolysate, which was prepared by hydrolysing wood chips using concentrated sulfuric acid.

     strain for recombinant DNA manipulation in this study was NovaBlue {endA1 hsdR17(rK12mK12+) supE44 thi-1 gyrA96 relA1 lac recA1/F?ä [proAB+lacIqZ??M15::Tn10 (Tetr)]} (Novagen, Madison, Wisconsin, USA). S. cerevisiae MT8-1 (MATa ade his3 leu2 trp1 ura3) was used for cell-surface expression and fermentation (Tajima et al. 1985). E. coli was grown in Luria?CBertani medium (10 g/l tryptone, 5 g/l yeast extract, 5 g/l sodium chloride) containing 100 mg/l ampicillin. After precultivation, the yeast was aerobically cultivated at 30?ãC in synthetic medium [SD medium; 20 g/l glucose and 6.7 g/l yeast nitrogen base without amino acids (Difco Laboratories, Detroit, Michigan, USA) with appropriate supplements] containing 20 g/l casamino acids (SDC medium, Difco). Lignocellulosic hydrolysate The wood chip hydrolysate used was provided by the JGC (Yokohama, Japan). This hydrolysate was prepared by the modified concentrated sulfuric acid hydrolysis method, which was developed by Arkenol (Irvine, California, USA) (Farone and Cuzens 1997, 1998). The resulting hydrolysate was adjusted to pH 7.0 using CaCO3, and used as lignocellulose hydrolysate for fermentation performance analysis. The wood chip hydrolysate contained 57.5 g/l glucose, 15.1 g/l mannose, 9.3 g/l galactose, 11.2 g/l xylose, and 10.2 g/l cellooligosaccharides.

     Materials and methods

     Strains and media All the microbial strains constructed and used in this study are listed in Table 1. The Escherichiacoli strain used as host

     Table 1 Characteristics of strains and plasmids used in this study Strain or plasmid Yeast strains S. cerevisiae MT8-1 S. cerevisiae MT8-1/Con MT8-1/Xyl MT8-1/ BGL Bacterial strains E. coli NovaBlue Relevant feature Source or reference

     MATa ade his3 leu2 trp1 ura3

     Tajima et al. (1985) This study This study This study

     No expression (control strain) Expressing XR, XDH, and XK Displaying BGL1

     endA1 hsdR17(rK12-mK12+)supE44 Novagen thi-I gyrA96 relA1 lac recA1/F?ä[proAB + lacIqZ??M15::Tn10(Tetr)] Stratagene Stratagene This study This study

     Plasmids pRS403 HIS3 no expression (control plasmid) pRS406 URA3 no expression (control plasmid) pIUX1X2XK URA3 intracellular coexpression of XYL1, XYL2, and XKSI genes pIBG13 HIS3 surface expression A. acleaus ?Â-glucosidase gene

     Fig. 1 Physical maps of pIUX1X2XK for intracellular coexpression

    of XYL1, XYL2, and XKS1 (a), and expression plasmid pIBG13 for yeast cell surface displaying BGL1-?Á-agglutinin fusion protein (b). s.s. secretion signal sequence of Rhizopusoryzae glucoamylase gene

     1138

     Construction of plasmids for cell surface display and intracellular expression The plasmid pIUX1X2XK (Fig. 1a) used for the intracellular coexpression of the XYL1, XYL2, and XKS1 genes was constructed as follows. The following fragments were prepared by polymerase chain reaction (PCR) using primers and plasmids as described by the previous study (Katahira et al. 2004): a 2.2-kbp KpnI-XhoI DNA fragment composed of the glyceraldehyde-3-phosphate dehydrogenase (GAPDH) promoter, XYL1 gene, and the GAPDH terminator; a 2.3-kbp XhoI-NotI DNA fragment composed of the GAPDH promoter, XYL2 gene, and the GAPDH terminator; and a 3.04-kbp NotI-SacII DNA fragment composed of the GAPDH promoter, the XKS1 gene, and the GAPDH terminator. These DNA fragments were introduced into the KpnI-XhoI section, XhoI-NotI section, and NotI-SacII section of the plasmid pRS406 (Stratagene, La Jolla, California, USA), and the resulting plasmid was designated pIUX1X2XK. The plasmid pIBG13 for the cell surface display of A. aculeatus No. F-50 BGL1 was constructed as follows. The 2.5-kbp NcoI-XhoI DNA fragment encoding bgl1 gene was prepared by PCR performed with primers 5?ä-GATCTCC

    ATGGCTGATGAACTGGCGTTCTCTCCTCCTTTC-3?ä and

    5?ä-TGGCGCTCGAGCCTTGCACCTTCGGGAGC GCCGCGTGAAG-3?ä and with the plasmid pBG211 as template. This DNA fragment was digested with NcoI and XhoI and introduced into the NcoI-XhoI site of the cellsurface expression plasmid pIHCS (Fujita et al. 2002) containing the genes encoding the secretion-signal sequence of the glucoamylase gene from Rhizopus oryzae and the 3?ä-half region of the ?Á-agglutinin gene (Lipke et al. 1989). The resulting plasmid was designated pIBG13 (Fig. 1b). Yeast transformation Yeast transformation was carried out by the lithium acetate method using the YEASTMAKER yeast transformation system (Clontech Laboratories, Palo Alto, California, USA). The plasmids pRS406 and pIUX1X2XK were digested with NcoI and PstI, respectively, within the URA3 gene, and the plasmids pRS403 and pIBG13 were digested with NdeI within the HIS3 gene. Then, the linear products were used to transform S. cerevisiae MT8-1. The resulting yeast strains are summarized in Table 1. Enzyme assays Cell extracts for the assays of xylose metabolic enzymes were prepared as follows: After cultivation in SDC medium for 48 h at 30?ãC, cells were harvested by centrifugation at 6,000?Ág for 10 min at 4?ãC and washed with distilled water three times. Then, the cells were resuspended in 0.1 M sodium phosphate buffer with an equal volume of glass beads (0.5 mm diameter). The cells were disrupted for

     15 min using 0.5 mm of glass beads. The lysate was centrifuged at

    14,000?Ág for 5 min at 4?ãC, and the supernatant was analyzed for enzyme activities. Protein concentration was measured according to the Bradford method (Bradford 1976), with bovine serum albumin (BSA) as standard. XR activity was measured spectrophotometrically by monitoring NADPH oxidation at 340 nm (Smiley and Bolen 1982; Walfridsson et al. 1997) in a reaction mixture with the following composition: 10 mM sodium phosphate buffer (pH 7.0) at 30?ãC, 0.2 M xylose, and 0.12 mM NADPH as substrate. XDH activity was measured spectrophotometrically by monitoring nicotinamide adenine dinucleotide (NAD+) reduction at 340 nm (Smiley and Bolen 1982; Walfridsson et al. 1997) in a reaction mixture with the following composition: 0.1 M Tris?CHCl (pH 7.0) at 30?ãC, 50 mM xylitol, and 2 mM NAD+ as substrate. The XK reaction forming adenosine diphosphate was coupled with pyruvate kinase (PK) and lactate dehydrogenase (LDH) reactions. XK activity was measured

    spectrophotometrically by monitoring NADH oxidation by LDH at 340 nm (Shamanna and Sanderson 1979; Eliasson et al. 2000) in a reaction mixture of the following composition: 0.1 M Tris?CHCl (pH 7.0), 2 mM MgCl2, 8 mM NaF, 2 mM ATP, 0.2 mM phosphoenolpyruvate, 3 mM reduced glutathione, 10 U of LDH, 10 U of PK, 0.1 mM NADH, and 8.5 mM xylulose. One unit of enzyme activity was defined as the amount of enzyme that released 1 ?Ìmol each of NADPH and NADH reduced or oxidized from the substrate per minute. ?Â-glucosidase activity was measured as described previously (Fujita et al. 2002) with 50 mM sodium acetate buffer (pH 5.0) at 30?ãC with 1 mM p-nitrophenyl-?Â-Dglucopyranoside (pNPG) (Nacalai Tesque, Kyoto, Japan) as substrate. For activity assay, cultured yeast cells, were harvested by centrifugation at 6,000?Ág for 10 min at 4?ãC and washed with distilled water three times. The optical density at 600 nm (OD600) of the reaction mixture was adjusted to 1. After the reaction, the supernatant was separated by centrifugation at 14,000?Ág for 5 min at 4?ãC, and the amount of p-nitrophenol released was determined by measuring absorbance at 400 nm. One unit of enzyme activity was defined as the amount of enzyme that released 1 ?Ìmol of p-nitrophenol from the substrate per minute. Analysis of substrates and products Glucose, xylose, mannose, galactose, glycerol, and xylitol in the fermentation medium were analyzed by high performance liquid chromatography (HPLC) (Shimadzu, Kyoto, Japan). A Shim-pack SPR-Pb column (Shimadzu) was used together with a refractive index detector (model RID-10A, Shimadzu). The HPLC system was operated at 80?ãC with water (flow rate, 0.4 ml/min) as mobile phase. Ethanol concentration was measured bygas chromatography. A chromatograph (model GC-8A; Shimadzu) fitted with a flame ionization detector was operated under the following conditions: glass column (3.2 mm?Á2.0 m)

     1139

     packed with Thermon-3000 (Shimadzu); temperature of column,

    injector, and detector, 180?ãC; and nitrogen carrier gas flow rate, 25 ml/min. The concentration of cellooligosaccharides in the lignocellulose hydrolysate was determined as described below. The lignocellulose hydrolysate was hydrolyzed by commercially available cellulase and ?Â-glucosidase (Sigma Chemical, St. Louis, MO, USA) in a reaction mixture with 50 mM sodium acetate buffer (pH 5.0) at 30?ãC. After 12 h incubation, the hydrolysate was analyzed by HPLC, as described above. The concentration of cellooligosaccharides in the

    lignocellulose hydrolysate was derived from the increase in glucose concentration. Fermentation The yeast transformants were aerobically cultivated in SDC medium for 48 h at 30?ãC. Their cells were collected by centrifugation at 3,000?Ág for 10 min at 4?ãC and washed with distilled water three times. Then, the cells were inoculated into the fermentation medium containing 6.7 g/l yeast nitrogen base without amino acids, 20 g/l casamino acids, and varieties of sugars as the sole carbon source. All fermentations were performed at 30?ãC with mild agitation at 100 rpm in 100-ml closed bottles equipped with a bubbling CO2 outlet. The initial cell density in the fermentation medium was adjusted to an OD600 of 20. During fermentation, cell growth was observed by monitoring optical density at 600 nm. Dry cell weight (DW) was determined as follows: the culture broth (10 ml) adjusted to various OD600 values was pelleted in preweighed 15-ml tubes by centrifugation at 3,000?Ág for 10 min. After resuspension in 10 ml of distilled water and further centrifugation, the pellets were freeze-dried with FreeZone FZ-1 (Labconco, Kansas, MO, USA). DW was calculated by reweighing the tubes. Cell concentration was estimated from the OD600-to-DW correlation (0.31 g-DW/ OD600 values). Specific ethanol production rate was determined as the coefficient of the linear regression of ethanol concentration vs the time integration of biomass concentration.

     BGL showed no activity. The XR/XDH ratios of MT8-1/ Xyl and MT8-1/Xyl/BGL were 0.10 and 0.13, respectively. The BGL1 activity of the yeast strains displaying BGL1 was measured using both culture supernatant and cell pellet fractions. As shown in Table 2, the strains displaying BGL1 (MT8-1/BGL and MT8-1/Xyl/BGL) showed similar BGL activities [15.44 and 13.46 U/g (dry weight) of cells, respectively], while MT8-1/Con and MT8-1/Xyl showed no BGL1 activity. No BGL1 activity was detected in culture supernatant for all the strains (data not shown). Ethanol fermentation from xylose and cellobiose Ethanol fermentation from xylose was performed with 50 g/l xylose as carbon source using MT8-1/Xyl, MT8-1/ Xyl/BGL, and MT8-1/Con (Fig. 2). The xylose-utilizing MT8-1/Xyl and MT8-1/Xyl/BGL produced about 18 g/l ethanol, and xylose was consumed at about 47 g/l at 72 h; no ethanol production or xylose consumption was observed in the control strain. The specific ethanol production rates of MT8-1/Xyl and MT8-1/Xyl/BGL were 0.064 and 0.058

    g of ethanol per gram of DW per hour, respectively. The ethanol yield estimated from the grams of ethanol produced per gram of consumed xylose after 72 h fermentation by both MT8-1/Xyl and MT8-1/Xyl/BGL was 0.37 g per gram (Table 3). This ethanol yield corresponds to 72.5% of the theoretical yield (0.51 g of ethanol per gram of xylose). Xylitol accumulation in fermentation medium was maintained at a level lower than 2 g/l in both MT8-1/ Xyl and MT8-1/Xyl/BGL (data not shown). Meanwhile, glycerol accumulation was observed in both strains (about 6 g/l at a maximum; data not shown). These yeast strains hardly grew during the fermentation experiments. The pHs of the fermentation medium of MT8-1/Xyl and MT8-1/ Xyl/BGL decreased from 5.6 to 4.6 after 72 h fermentation. Ethanol fermentation from cellobiose was performed with 50 g/l cellobiose as carbon source using the yeast

     Result

     Enzyme activities In this study, we generated the xylose-fermenting yeast strain (MT8-1/Xyl), yeast strain displaying BGL1 on its cell surface (MT8-1/BGL), and xylose-fermenting and BGL1-displaying yeast strain (MT8-1/Xyl/BGL) by integratiing pIUX1X2XK for the coexpression of XYL1, XYL2, and XKS1, and pIBG13 for displaying BGL1 (Table 1). The XR, XDH, and XK activities of all yeast strains were measured (Table 2). MT8-1/Xyl and MT8-1/Xyl/ BGL with the three genes for xylose metabolism introduced into their chromosome showed similar XR, XDH, and XK activities, while MT8-1/Con and MT8-1/

     Fig. 2 Time course of ethanol fermentation from 50 g/l xylose as sole carbon source by recombinant yeast strains. Triangle MT8-1/ Con; circle MT8-1/Xyl; and square MT8-1/Xyl/BGL. Open and closedsymbols show the xylose and ethanol concentrations, respectively. The data points are the average of two independent experiments

     1140 Table 2 Distribution of XR, XDH, and XK activities in cell extracts of cells, and BGL1 activity of cells Strain Enzyme activity BGL1 activity in cells (U/mg of protein)a [U/g (dry weight) of cells]a XR MT8-1/Con MT8-1/Xyl MT8-1/BGL1 MT8-1/Xyl/BGL1

     a

     XDH XK ND 1.83 ND 1.75 ND 0.39 ND 0.37 ND ND 15.44 13.46

     NDb 0.19 ND 0.23

     Values are average of three independent experiments. Derivations are always below 10% of the average b ND not detected Fig. 3 Time course of ethanol fermentation from 50 g/l cellobiose as sole carbon source by recombinant yeast strains. Triangle MT8-1/ Con; diamond MT8-1/BGL; and square MT8-1/Xyl/BGL. Open and closed symbols show the cellobiose and the ethanol concentrations, respectively. The data points are the average of two independent experiments

     strains displaying BGL1 on their cell surface, MT8-1/BGL and MT8-1/Xyl/BGL, and the control strain MT8-1/Con (Fig. 3). MT8-1/BGL

    and MT8-1/Xyl/BGL consumed cellobiose completely within 24 h and produced final ethanol concentrations of 19.3 and 18.4 g/l, respectively, whereas MT8-1/Con showed neither cellobiose reduction nor ethanol production in the medium. MT8-1/BGL and MT8-1/Xyl/BGL showed specific ethanol production rates of 0.10 and 0.08 g g1 h1, respectively (Table 3). The ethanol yields from consumed cellobiose after 24 h fermentation by MT8-1/BGL and MT8-1/Xyl/BGL were 0.37 and 0.38 g per gram, respectively (Table 3). These ethanol yields correspond to 72.5 and 74.5% of the theoretical yield (0.51 g of ethanol per gram of glucose), respectively. Glucose accumulation was hardly observed during the fermentation (data not shown). Meanwhile, glycerol accumulation was observed in both strains (about 4 g/l at maximum; data not shown). Additionally, during the fermentation experiments, these strains grew gradually, and cell concentrations reached approximately 10 g/l DW after 24 h fermentation. The pHs of the fermentation medium of MT8-1/BGL and MT8-1/Xyl/BGL decreased from 5.6 to 4.3 and 4.2, respectively, after 24 h fermentation.

     Ethanol fermentation from sugar mixture A model sugar mixture reflecting the sugar composition of the lignocellulose hydrolysate was fermented by MT8-1/ Xyl/BGL and MT8-1/Con. The fermentation medium contained 50 g/l glucose, 10 g/l mannose, 5 g/l galactose, 10 g/l xylose, and 10 g/l cellobiose. Glucose and mannose were rapidly fermented within 8 h by both strains. The recombinant strain MT8-1/Xyl/BGL degraded cellobiose and metabolized xylose (Fig. 4a), whereas the control strain utilized neither cellobiose nor xylose (Fig. 4b). MT81/Xyl/BGL metabolized 7.3 g/l xylose after 36 h fermentation, and completely hydrolyzed cellobiose within 12 h. Then, the highest ethanol concentration produced by MT81/Xyl/BGL reached approximately 32.5 g/l after 36 h fermentation, whereas the ethanol production by the control strain was less than 26 g/l after 16 h fermentation. MT8-1/Xyl/BGL produced 0.83 g/l xylitol at 36 h (data not shown). Meanwhile, galactose was slightly consumed by these strains. After 36 h fermentation, galactose remained at about 3 g/l in the medium of these strains (data not shown). This phenomenon suggests that the strain MT8-1 used in this study has a low galactose metabolic performance. MT8-1/Xyl/BGL and MT8-1/Con showed specific ethanol production rates of 0.41 and 0.40 g g1 h1, respectively (Table 4). The ethanol yields of MT8-1/Xyl/ BGL and MT8-1/Con estimated from the grams of ethanol produced per gram of total sugar added at the beginning were 0.39 and 0.32 after 36 h fermentation, respectively (Table 4). During the fermentation experiments, these strains grew gradually and cell concentrations of MT8-1/ Con and MT8-1/Xyl/BGL reached 10.8 g and 11.2 g/l DW after 24 h fermentation, respectively (data not shown). The pHs of the fermentation medium of MT8-1/Con and MT81/Xyl/BGL decreased from 5.6

to 4.1 and 4.2, respectively, after 24 h fermentation.

     Table 3 Specific ethanol production rates and ethanol yields of recombinant strains in fermentation of xylose and cellobiosea Strain Specific ethanol production rateb Xylose MT8-1/Con MT8-1/Xyl MT8-1/BGL MT8-1/Xyl/BGL

     a

     Ethanol yieldc Xylose Cellobiose 0 0.37 0 0.37 0 0 0.37 0.38

     Cellobiose 0 0 0.10 0.08

     0 0.06 0 0.06

     Displayed values are the average of two independent experiments Deviations are always below 10% of the average b Specific ethanol production rate is expressed in grams of ethanol/ grams of dry weight of cells per hour c Ethanol yeild is expressed in grams of produced ethanol/grams of consumed sugar after 72 h of fermentation in xylose and 24 h of fermentation

     1141

     Fig. 4 Fermentation of sugar mixture by MT8-1/Xyl/BGL (a) and MT8-1/Con (b). Closed square ethanol; open triangle glucose; open circle xylose; open diamond cellobiose; and cross mannose. The data points are the average of two independent experiments

     Fig. 5 Fermentation performances of MT8-1/Xyl/BGL (a) and MT8-1/Con (b) from lignocellulosic hydrolysate as carbon source. Closed square ethanol; open triangle glucose; open circle xylose; and cross mannose. The data points are the average of two independent experiments

     Ethanol fermentation from lignocellulose hydrolysate The lignocellulose hydrolysate was fermented by MT8-1/ Xyl/BGL and MT8-1/Con. The fermentation medium contains 40.2 g/l glucose, 10.5 g/l mannose, 6.5 g/l galactose, 7.9 g/l xylose, and 7.2 g/l cellooligosaccharides. In the fermentation of the lignocellulose hydrolysate, glucose and mannose were rapidly fermented within 8 h by both strains (Fig. 5a,b). In the fermentation by MT8-1/Con

     Table 4 Specific ethanol production rates and ethanol yields of recombinant strains in fermentation of sugar mixture and lignocellulosic hydrolysatea Strain Specific ethanol production rateb Sugar mixture MT8-1/Con MT8-1/Xyl/ BGL

     a

     Ethanol yieldc Hydrolysate

     as reference strain, the highest ethanol concentration was about 22 g/l after 12 h fermentation (Fig. 5b). Meanwhile, MT8-1/Xyl/BGL completely hydrolyzed and assimilated cellooligosaccharides and assimilated about 6 g/l xylose within 36 h, and achieved the highest ethanol concentration of 30.3 g/l after 36 h fermentation (Fig. 5a). A small amount of galactose was fermented within 36 h (data not shown). In the fermentation media of MT8-1/Xyl/BGL and the control strain,

    xylitol accumulation was hardly observed during the experiment (less than 1 g/l; data not shown). MT8-1/Xyl/BGL and MT8-1/Con showed specific ethanol production rates of 0.42 and 0.40 g g1 h1, and ethanol yields of 0.41 and 0.31 g per gram, respectively (Table 4).

     Hydrolysate Sugar mixture 0.40 0.42 0.32 0.39

     Discussion

     0.31 0.41

     0.40 0.41

     Displayed values are the average of two independent experiments. Deviations are always below 10% of the average b Specific ethanol production rate is expressed in grams of ethanol/ grams of dry weight of cells per hour c Ethanol yield is expressed in grams of produced ethanol per grams of total sugar after 36 h of fermentation

     We generated the recombinant yeast strains with xyloseassimilating and cellooligosaccharide-degrading abilities by integrating the genes for XYL1, XYL2, XKS1, and BGL1-?Á-aggulutinin in the yeast chromosome. Then, the fermentation performance of these recombinant strains was examined using xylose, cellobiose, the model sugar mixture, and the lignocellulose hydrolysate.

     1142

     MT8-1/BGL and MT8-1/Xyl/BGL showed similar BGL1 activities. In its fermentation, cellobiose was completely hydrolyzed within 24 h by both strains. In addition, both strains showed similar specific ethanol production rates and ethanol yields. Meanwhile, in glucose fermentation, specific ethanol production rates were approximately 0.43 g g1 h1 for all the strains used in this study (data not shown). Specific ethanol production rate in fermentation of cellobiose was a quarter of that in the glucose fermentation. These results indicate that cellobiose hydrolysis by BGL1 is the rate-limiting step of ethanol fermentation from cellobiose. Thus, further work is needed to improve ?Â-glucosidase activity for the enhancement of ethanol production rate. The ethanol fermentation from xylose by these recombinant strains was performed efficiently without a marked xylitol accumulation, although a small amount of glycerol was accumulated (data not shown). As shown in Table 3, the ethanol yields of the xylose-assimilating strains (MT81/Xyl and MT8-1/Xyl/BGL) were higher than those of TMB3001 and 1400(pLNH32) (0.31 and 0.3, respectively), well-known xylose-fermenting S. cerevisiae strains (Ho et al. 1998; Eliasson et al. 2000; Jeppsson et al. 2002). The higher ethanol yields of MT8-1/Xyl and MT8-1/Xyl/ BGL must be due to a very low xylitol accumulation (lower than 2 g/l). The low XR/XDH activity ratios of MT8-1/Xyl and MT8-1/Xyl/BGL (0.10 and 0.13) may contribute to a low xylitol accumulation (Walfridsson et al. 1997). Glycerol accumulation may be caused by the reoxidation of NAD+, produced via NAD-dependent XDH, with glycerol formation, because of the

Report this document

For any questions or suggestions please email
cust-service@docsford.com