Influence of microbial cultivation during bio-reduction with in-situ product crystallization (ISPC)

An in-situ product crystallization process (ISPC) was developed for a crystalline product formed during bioreduction coupled with biocatalyst cultivation. The model reaction was the asymmetric reduction of 4-oxoisophorone (OIP) by baker’s yeast (Saccharomyces cerevisiae). Yeast cells were cultivated fed-batch in the reactor with an initial cell concentration (CXi) of 1 gdw.L to reach a maximum concentration of 30 gdw.L. The desired product, 6R-dihydro-oxoisophorone (DOIP), is also degraded by baker’s yeast mainly to an unwanted by-product (4S,6R-actinol); thus, it was removed immediately from the fermentor via an external crystallization unit in the integrated process. The OIP reduction rate was five times higher (≅ 0.33 mmol.gdw.h) with growing cells as compared to the reduction rate with resting cells. During the integrated process, OIP reduction was started when the optimum cell concentration was already reached in the reactor as the substrate (OIP) was found to inhibit cell growth at a constantly high concentration (COIP ≥ 55 mM) in the reactor. Final DOIP yield and selectivity were 85% and 99%, respectively. The product crystals were readily recovered, rod-like in shape, and tend to form aggregates. Typical DOIP crystals have an average diameter of 12 μm and length of 20 μm. INTRODUCTION Nowadays, the range of products that can be produced by fermentation (biotransformation) is increasing as productivity improves dramatically using protein and metabolic engineering. In addition, maximizing the biocatalyst concentration in the reactor and/or immediately removing inhibiting or degrading products during the process can also raise bioreactor productivity. The former requires cell cultivation during the process especially when the biocatalyst is not commercially or readily available and the latter involves in-situ product removal [Lye and Woodley, 1999; Stark et al., 2003; Buque-Taboada et al., 2004a]. This work aims to demonstrate experimentally the feasibility of implementing cell cultivation concomitant with bio-reduction and in-situ product crystal formation in an integrated fermentationcrystallization process. In-situ product crystallization is considered in this work as it can directly provide the desired product (in already pure form) without the need for an auxiliary phase [van der Wielen and Luyben, 1992; Buque-Taboada et al., 2004a]. This approach can be generally applicable in processes, especially those involving cell cultivation prior to or simultaneously with the biocatalytic formation of the product and its subsequent crystallization. This process strategy is also widely suitable for processes where it is required to control and reduce the product concentration in the reactor in order to prevent product toxicity, inhibition and/or degradation, and in addition, simplify the product separation and recovery steps. The chosen model reaction was the asymmetric reduction of 4-oxoisophorone (OIP) using baker’s yeast (Saccharomyces cerevisiae) as biocatalyst (Figure 1). The desired product is known as 6Rdihydro-oxoisophorone (DOIP), which is a key intermediate in carotenoid synthesis [Leuenberger, 1985] and in the production of saffron and tobacco flavours [Sode et al., 1987]. As baker’s yeast is known to also degrade DOIP mainly to 4S,6R-actinol, an unwanted by-product [Leuenberger et al., 1976] , this must be removed from the fermentor as soon as it is formed to prevent low product yield and selectivity. In this case, in-situ product crystallization (ISPC) is appropriately applied. In the previous work [Buque-Taboada et al., 2004a], it was shown that with resting cells of S. cerevisiae, the integrated process was the most efficient compared to batch and fed-batch alternative configurations. Fed-batch processing might also be favourable for process systems where the biocatalyst needs to be cultivated in the reactor prior to or concomitant with bio-reduction and product recovery. Although baker’s yeast is not an exceptional organism as it is cheap, readily available, and thus, does not need to be cultivated in the reactor, it is an interesting case to perform reductions with precultivated yeast cells. In many instances, using whole cells as biocatalyst would prove economically favourable as this allows for an easy and in-vivo cofactor regeneration in the cell, which sustains catalytic activity for redox reactions. Aeration and nutrient feeding can, however, be the most important constraints to consider, and can potentially complicate the whole biocatalytic process.