In this study, two statistical experimental designs, Plackett-Burman design (PBD) and response surface methodology (RSM), were employed to enhance the expression of 2,3-dihydroxybiphenyl 1,2-dioxygenase (BphC_LA-4), which was subsequently used for the construction of catechol biosensor. Ten important factors were evaluated by PBD, and four significant parameters were then optimized by RSM. Under the favorable fermentation conditions, the maximal specific activity of BphC_LA-4 was about 0.58 U/mg with catechol as substrate. Meanwhile, homology modeling and molecular docking were utilized to help understand the interaction between BphC_LA-4 and catecholic substrates, which illustrated that BphC_LA-4 presented lower binding affinity towards 4-methylcatechol in comparison with 3-methylcatechol and catechol. Interestingly, the BphC_LA-4 enzyme electrode prepared by SiO(2) sol-gel showed good response to all these three catecholic compounds. The differences of selectivity to 4-methylcatechol between free and immobilized enzyme implied that the introduction of electro-catalysis might have an effect on the enzyme-catalysis process. (C) 2011 Elsevier Ltd. All rights reserved.