Because the use of basidiomycetes in bioethanol production is not common, and the ethanol fermentation abilities of basidiomycetes are not well characterized, we investigated the properties of ethanol fermentation by F. velutipes to determine its suitability for CBP (Mizuno et al., 2009b). Before the experiment, to obtain a suitable strain for CBP, 10 F. velutipes strains, culture stock of the Forest Institute of Toyama Prefectural Agricultural, Forestry, and Fisheries Research Center, were screened for cellulase production and ethanol fermentation. The Fv-1 strain was selected for further study because it not only produces high levels of cellulases, but also because its ability to ferment ethanol is superior to the other strains.

Firstly, fermentation of D-glucose was done by F. velutipes Fv-1. Figure 1A shows a conversion of 1% w/v of D-glucose to ethanol by F. velutipes. The consumption of D-glucose started gradually after incubation, and it was depleted after 6 d. Ethanol production correlated with sugar consumption, and it reached a maximum after 6 d. Thereafter, the amount of ethanol decreased gradually. Finally, F. velutipes converted 10 g/l of D-glucose to 4.5 g/l of ethanol, equivalent to a theoretical ethanol recovery rate of 88%. In the case of ethanol production from 5% w/v D-glucose, ethanol production reached a maximum, and all of the D-glucose was consumed after 18 d of incubation (Fig. 1B), and 50 g/l of D-glucose was converted to 22.4 g/l of ethanol, equivalent to a theoretical ethanol recovery rate of 87%. The conversion rate was the same as the case of 1% w/v of D-glucose. Because the incubation time to ferment 1% w/v sugar is shorter than the case of 5% w/v, we employed 1% w/v of sugar concentration in subsequent experiments.

Secondary, determination of the fermentation specificity of sugars by F. velutipes Fv-1 was done using various monosaccharides. As shown in Fig. 2, both D-mannose and D-fructose were converted to ethanol by F. velutipes. Consumption of D-mannose occurred slightly faster than that of D-glucose; it started immediately after incubation and was completely depleted after 5 d. Ethanol production from D-mannose was similar to that from D-glucose. It started during the first day of incubation and reached a maximum after 6 d. Furthermore, 4.4 g/l of ethanol was produced from 10 g/l of D-mannose, equivalent to a theoretical ethanol recovery rate of 86% (Fig. 2A). In contrast, consumption of D-fructose was slower than that of D-mannose. It started slowly after incubation and took 7 d to completely consume the D-fructose. Production of ethanol correlated with sugar consumption, and

maximum conversion of D-fructose to ethanol was observed after 6 d. Upon completion of incubation, 4.0 g/l of ethanol was obtained from 10 g/l of D-fructose (Fig. 2B), yielding a theoretical conversion rate of 77%. In contrast to these sugars, F. velutipes did not convert L — arabinose, D-xylose, or D-galactose to ethanol (Figs. 2C, 2D, and 2E). Although there was slight consumption of D-xylose and D-galactose during incubation, ethanol production was not observed. In the case of L-arabinose, little sugar consumption was observed.

Next, we examined the fermentation specificity of F. velutipes Fv-1 toward various disaccharides. As shown in Fig. 3, F. velutipes possibly converted these sugars to ethanol and produced high yields. The theoretical conversion rates of these sugars were 83% and 77% from sucrose and maltose respectively. Degradation of sucrose was observed immediately after the incubation to import the sugar. The amount of reducing sugars was maximum on day 3 and was completely consumed after 7 d of incubation. Ethanol production was observed 1 d after incubation, and the amount of ethanol reached a maximum after 6 d. Finally, 4.5 g/l of ethanol was produced from 10 g/l of sucrose (Fig. 3A). In the case of maltose, degradation was observed on the first day of incubation, and the amount of reducing sugars reached a maximum after 2 d. Furthermore, the reducing sugars were completely depleted after 7 d of incubation. Ethanol production started during the first day of incubation and reached a maximum after 7 d. At the end of incubation, 10 g/l of maltose was converted to 3.8 g/l of ethanol (Fig. 3B). No conversion of xylobiose to ethanol was detected (data not shown), but a significant amount of ethanol production was observed when cellobiose was used as the carbon source (Fig. 4A). Cellobiose began degrading during the first day of incubation, and both D-glucose and cellobiose were completely depleted after 8 d. p-Glucosidase activity increased gradually during incubation. Ethanol production started after 1 d of incubation, and the amount of ethanol reached a maximum after 8 d. Upon completion of incubation, 10 g/l of cellobiose was converted to 4.5 g/l of ethanol (Fig. 4A). The theoretical conversion rate was 83%, a value similar to that of glucose and significantly higher than that of maltose. A high yield of ethanol was observed also in the higher concentration of cellobiose (Fig. 4D). Finally, 25 g/l of ethanol was produced from 50 g/l of D-glucose, and the theoretical conversion rate was 91%.

Since cellobiose was converted to ethanol at a relatively high rate, the conversions of cello- oligosaccharides to ethanol by F. velutipes were also investigated. Figures 4B and 4C show

the results of the conversion of cellotriose and cellotetraose to ethanol. Both cello- oligosaccharides were effectively converted to ethanol by F. velutipes. During incubation, cellotriose was initially hydrolyzed to D-glucose and cellobiose, and almost 80% of the initial amount of cellotriose was hydrolyzed by 2 d. Cellotriose was not detected after 5 d of incubation, and D-glucose and cellobiose were completely depleted after 7 d. p-Glucosidase

was slightly induced by 6 d, and the activity gradually increased after 6 d. The amount of ethanol increased during incubation and reached a maximum after 7 d of incubation. F. velutipes produced 4.2 g/l of ethanol from 10 g/l of cellotriose, equivalent to a theoretical conversion rate of 76% (Fig. 4B). In the case of cellotetraose, it was initially hydrolyzed to cellotriose, cellobiose, and D-glucose, and more than 90% of the cellotetraose was hydrolyzed by 2 d. Cellotetraose was not detected after 3 d of incubation, and cellotriose, cellobiose, and D-glucose were completely depleted after 4, 6 and 7 d respectively. p — Glucosidase activity increased rapidly over 2 d then decreased gradually from 2 d to 5 d, and stabilized at an activity level of about 30 mU / ml. The amount of ethanol increased after incubation, and 4.4 g/l of ethanol was produced from 10 g/l of cellotetraose after 7 d of incubation (Fig. 4C). The ethanol recovery for the theoretical conversion value was 78%.

To date, many microorganisms, including Saccharomyces cerevisiae, Zymmonas mobilis, Pichia stipitis, Rhizopus oryzae, and Clostridium thermocellum, have been reported to produce ethanol (DeMoss & Gibb, 1951; Maas et al., 2006; Ng et al., 1981; Parekh & Wayman, 1986; Weimer & Zeikus, 1977). In general, S. cerevisiae is the most widely used microorganism in the industry and is popular in bioethanol production, because it has high efficiency of ethanol production and high ethanol tolerance. However, we focused on basidiomycetes to develop CBP because these microorganisms have both lignocellulose degradation and ethanol fermentation abilities.

Here, we characterized properties of ethanol fermentation by F. velutipes Fv-1. The strain converted D-glucose to ethanol at a theoretical conversion rate of 88%, comparable to those of S. cerevisiae and Zymomonas (Swings & DeLey, 1977). On the other hand, F. velutipes scarcely converted pentose and D-galactose to ethanol (Fig. 2). These properties of F. velutipes are similar to those of S. cerevisiae (Barnett, 1976). Moreover, F. velutipes demonstrated the preferable features for CBP when oligosaccharides were used as starting materials (Figs. 3 and 4). The tested oligosaccharides were converted to ethanol at almost the same rate as that of D-glucose, and p-glucosidase activity increased during fermentation. These features are indispensable in CBP, which requires saccharification and fermentation of cellulose contained in the cell wall. It has been reported that C. thermocellum and P. stipitis can ferment cellobiose (Parekh & Wayman, 1986). Furthermore, C. thermocellum can also convert cellulose to ethanol directly (Ng et al., 1981; Lynd et al., 1989; Weimer & Zeikus,

Fig. 4. Ethanol fermentation from (A) cellobiose, (B) cellotriose, (C) cellotetraose and (D) 5% cellobiose by F. velutipes

1997). However, this species cannot be used at the scene of ethanol production because fermentation of C. thermocellum is strongly inhibited at relatively low ethanol concentrations (5 g/l) (Herrero & Gomez, 1980). In contrast, it has been reported that basidiomycetes have tolerance of up to 120 g/l of ethanol (Okamura et al., 2001), and therefore basidiomycetes are more suitable for CBP than Clostridium strains. From these results, we concluded that F. velutipes possesses advantageous characteristics for use in CBP.