The use of whole cells is becoming a more common approach in pharmaceutical and agrochemical industries in order to obtain pure compounds with fewer production steps, higher yields, and cleaner processes, as compared to those achieved with traditional strategies Whole cells ate often used as enzymes pools, in particular when multi-step reactions and/or co-factor regeneration are envisaged. Nonetheless, published information on the scale-up of such systems both in aqueous and in two-phase aqueous-organic systems is relatively scat cc. The present work aims to evaluate suitable scale-up criteria in conventional and non-conventional medium for a whole-cell bioconversion that uses resting cells of Mycobacterium sp. NRRL B-3805 to cleave the side chain of beta-sitosterol, a poorly water-soluble substrate. The experiments were performed in 24-well microtiter plates and in 250 mL shaken flasks as orbital stirred systems, and in 300 mL stirred tanks as mechanically stirred systems. Results show that productivity yields were similar in all scales tested, when maintaining oxygen mass transfer coefficients constant in aqueous systems, or when maintaining constant volumetric power consumption in aqueous-organic two-phase systems Biotechnol. Bioeng. 2010;106: 619-626. (C) 2010 Wiley Periodicals, Inc.