Mixtures of selected thermostable enzymes (Table 2) were first evaluated for their hydrolytic efficiency by measuring the FPU activities at different tem­peratures (Fig. 3). The temperature optima of the new thermostable mixtures in the FPU activity assay were 5-10 °C higher than those of the commercial enzyme mixtures when a relatively short reaction time (60 min) in this assay was used. The relative FPU activity was set at the value of 100 at the refer­ence point at 50 °C. The maximum FPU activity of the novel enzyme mixture was about 25% higher at the optimum temperature at 65 °C as compared with the highest activity of the commercial reference enzyme at 60 °C. As could be expected, at lower temperature (35 °C), corresponding to the fermentation temperature of traditional yeasts in a simultaneous saccharification and fer­mentation process (SSF), the FPU activities of the thermostable preparations were slightly lower than those of the commercial T. reesei enzymes.

The thermostable enzyme mixture without added xylanase activity (TM 1) was evaluated on pure cellulose (Avicel) and compared with the commercial enzyme preparations (Celluclast supplemented with в-glucosidase) at 45 °C, 55 °C and 60 °C in a 48 h hydrolysis (Fig. 4). On pure cellulose, the mixture of thermostable enzymes gave nearly similar hydrolysis results at 60 °C as the T. reesei enzymes at 45 °C, i. e. thus enabling an increase of temperature of about 15 °C. At 60 °C, the hydrolysis yield of Avicel was about three — to four­fold better with the thermostable enzymes than with the commercial fungal enzymes. The highest hydrolysis yield was about 90% of the theoretical.

On the spruce substrate, the thermostable enzyme mixture resulted in an even more significant improvement in the performance at higher hydrolysis temperature as compared with the commercial enzymes. Thus, the hydrolysis yield was about threefold better at 55 °C and about fivefold better at 60 °C
using the thermostable enzyme mixture (Fig. 5). The hydrolysis was, however, also decreased with the thermoenzyme mixture at 60 °C. When comparing the hydrolytic performance of the commercial enzymes by increasing the temperature from 45 °C to 60 °C on Avicel and on spruce, it can be observed that the increased hydrolysis temperature decreased the performance on the natural lignocellulose substrate significantly more: from 70-10% on spruce, as compared with 90-30% on Avicel within 48 h. Obviously, the spruce sub­strate, even washed, contained compounds that, with increasing temperature, inhibited or inactivated not only the T. reesei enzymes, but also the ther­mostable enzymes.

High temperature enzyme mixtures suitable for hemicellulose-containing raw materials were evaluated in the hydrolysis of steam pretreated corn stover substrate (Fig. 6). With this raw material, the hydrolysis by the thermostable enzyme mixture at 45 °C was better than with the commercial preparation. The hydrolysis was still efficient at 55 °C and only slightly decreased at 60 °C with the thermostable enzyme mixture. The relative decrease of the hy­drolytic performance of both enzyme preparations was less pronounced on the corn stover substrate than with the spruce substrate at elevated tempera­tures. Based on HPLC analysis (Table 4) of the corn stover hydrolysates, the yield of glucose was around 90-95% of the theoretical after 72 h. The cor­responding yield of xylose was about 80-90% at temperatures up to 60 °C. The hydrolysis yields of the minor monosaccharide sugars, arabinose and galactose, were not significantly improved by the thermophilic enzyme mix­tures, indicating the absence of corresponding thermostable enzymes, i. e. arabinosidases and galactanases in the mixtures. In the hydrolysis of the

Table 4 Sugars released from steam pretreated spruce and corn stover (% of the initial sugar component in the substrate), analysed by HPLC

Enzymes Hydrolysis Sugars released Sugars released from corn stover

temp. from spruce

% of theoretical % of theoretical

(°C) Glucose Glucose Xylose Arabinose Galactose

Enzyme dosage was for reference enzymes: Celluclast 5 FPU g-1 substrate supplemented with 100 nkatg-1 Novozym 188; and for thermostable enzyme (TM 3) 5 FPU g-1 substrate. Hydrolysis time 72 h at pH 5, triplicates with mixing. Release of xylose, mannose and ara — binose from spruce substrate was below the reliable detection limit (less than 0.1% of the substrate) spruce substrate (Fig. 5, Table 4), only glucose was released. The individual sugar analyses corresponded well with the measured values of the reducing sugars.

9