Cyclodextrin glucanotransferase (CGTase) was produced when the Bacillus sp. TS1-1 was grown in a medium containing sago starch, yeast extract, phosphor-us and mineral salt sources, using shake flask mode at 37 degreesC for 24 h. Response surface methodology (RSM) was applied to optimize the medium constituents with respect to CGTase production and activity. A 2(4) full factorial design (first order model) was carried out to identify the significant effect of medium components towards CGTase production. The variables involved in this initial screening study were sago starch, yeast extract, K2HPO4 and MgSO4.7H(2)O. Statistical analysis of results have shown that only sago starch and yeast extract have a significant effect on CGTase production. A second-order model was proposed by using 2(2) central composite design to represent the production CGTase activity as a function of sago starch and yeast extract. The optimized values of 1.48% and 1.89% of sago starch and yeast extract was obtained, respectively. Under these proposed optimized conditions, the model predicted a CGTase activity of 79.66 U/ml and via experimental rechecking the model, an activity of 84 U/ml was attained. (C) 2004 Elsevier Inc. All rights reserved.