Fermentation of Cellulosic Substrates in Batch and Continuous Culture by Clostridium thermocellum

Publisher Website
Google Scholar
Abstract
Fermentation of dilute-acid-pretreated mixed hardwood and Avicel by Clostridium thermocellum was compared in batch and continuous cultures. Maximum specific growth rates per hour obtained on cellulosic substrates were 0.1 in batch culture and >0.13 in continuous culture. Cell yields (grams of cells per gram of substrate) in batch culture were 0.17 for pretreated wood and 0.15 for Avicel. Ethanol and acetate were the main products observed under all conditions. Ethanol:acetate ratios (in grams) were approximately 1.8:1 in batch culture and generally slightly less than 1:1 in continuous culture. Utilization of cellulosic substrates was essentially complete in batch culture. A prolonged lag phase was initially observed in batch culture on pretreated wood; the length of the lag phase could be shortened by addition of cell-free spent medium. In continuous culture with ∼5 g of glucose equivalent per liter in the feed, substrate conversion relative to theoretical ranged from 0.86 at a dilution rate (D) of 0.05/h to 0.48 at a D of 0.167/h for Avicel and from 0.75 at a D of 0.05/h to 0.43 at a D of 0.11/h for pretreated wood. At feed concentrations of D = 0.083/h. Lower conversion was obtained at higher feed substrate concentrations, consistent with a limiting factor other than cellulose. Free Avicelase activities of 12 to 84 mU/ml were observed, with activity increasing in this order: batch cellobiose, batch pretreated wood < batch Avicel, continuous pretreated wood < continuous Avicel. Free cellulase activity was higher at increasing extents of substrate utilization for both pretreated wood and Avicel under all conditions tested. The results indicate that fermentation parameters, with the exception of free cellulase activity, are essentially the same for pretreated mixed hardwood and Avicel under a variety of conditions. Hydrolysis yields obtained with C. thermocellum cellulase acting either in vitro or in vivo were comparable to those previously reported for Trichoderma reesei on the same substrates.