Trichoderma reesei expresses a large number of enzymes involved in lignocellulose hydrolysis and the mechanism of how these enzymes work together is too complex to study by traditional methods, e.g. by spiking with single enzymes and monitoring hydrolysis performance. In this study a multivariate approach, partial least squares regression, was used to see if it could help explain the correlation between enzyme profile and hydrolysis performance. Diverse enzyme mixtures were produced by Trichoderma reesei Rut-C30 by exploiting various fermentation conditions and used for hydrolysis of washed pretreated corn stover as a measure of enzyme performance. In addition, the enzyme mixtures were analyzed by liquid chromatography - tandem mass spectrometry to identify and quantify the different proteins. A multivariate model was applied for prediction of enzyme performance based on the combination of different proteins present in an enzyme mixture. The multivariate model was used for identification of candidate proteins that are correlated to enzyme performance on pretreated corn stover. A very large variation in hydrolysis performance was observed and this was clearly caused by the difference in fermentation conditions. Besides β-glucosidase, the multivariate model identified several xylanases, Cip1 and Cip2 as relevant proteins to study further.
- Trichoderma reesei
- Liquid chromatography–tandem mass spectrometry
- Mathematical modeling