Xylitol production by immobilized recombinant Saccharomyces cerevisiae in a continuous packed‐bed bioreactor

Abstract

Continuous xylitol production with two different immobilized recombinant Saccharomyces cerevisiae strains (H475 and S641), expressing low and high xylose reductase (XR) activities, was investigated in a lab-scale packed-bed bioreactor. The effect of hydraulic residence time (HRT; 1.3–11.3 h), substrate/cosubstrate ratio (0.5 and 1), recycling ratio (0, 5, and 10), and aeration (anaerobic and oxygen limited conditions) were studied. The cells were immobilized by gel entrapment using Ca-alginate as support and the beads were treated with Al3+ to improve their mechanical strength. Xylose was converted to xylitol using glucose as cosubstrate for regeneration of NAD(P)H required in xylitol formation and for generation of maintenance energy. The stability of the recombinant strains after 15 days of continuous operation was evaluated by XR activity and plasmid retention analyses. Under anaerobic conditions the volumetric xylitol productivity increased with decreasing HRT with both strains. With a recycling ratio of 10, volumetric productivities as high as 3.44 and 5.80 g/L · h were obtained with the low XR strain at HRT 1.3 h and with the high XR strain at HRT 2.6 h, respectively. However, the highest overall xylitol yields on xylose and on cosubstrate were reached at higher HRTs. Lowering the xylose/cosubstrate ratio from 1 to 0.5 increased the overall yield of xylitol on xylose, but the productivity and the xylitol yield on cosubstrate decreased. Under oxygen limited conditions the effect of the recycling ratio on production parameters was masked by other factors, such as an accumulation of free cells in the bioreactor and severe genetic instability of the high XR strain. Under anaerobic conditions the instability was less severe, causing a decrease in XR activity from 0.15 to 0.10 and from 3.18 to 1.49 U/mg with the low and high XR strains, respectively. At the end of the fermentation, the fraction of plasmid bearing cells in the beads was close to 100% for the low XR strain; however, it was significantly lower for the high XR strain, particularly for cells from the interior of the beads. © 1996 John Wiley & Sons, Inc.