Indian Journal of Experimental Biology Vol. 42, November 2004, pp. 1123-1131 Formulation of medium constituents by multiresponse analysis of central composite design to enhance chitinase production in Pantoea dispersa Vipul Gohel 3 , Smita Trivedib, Pranav Vyas" & H S Chhatpar a <*l "Department of Microbiology and Biotechnology Centre, Faculty of Science, M.S. University of Baroda, Vadodara 390 002, India bDepartment of Statistics, Faculty of Science, M.S.University of Baroda, Vadodara 390 002, India Received 26 April 2004; revised 9 July 2004 In the present study, a high chitinase producing strain Pa ntoea dispersa was isolated from the sea dumps at Bhavnagar, Indi a. Chitin, urea, CaC1 2 and MgS0 4 7H 2 0 were variables used in central composite design for chitinase production. Chitinase, biomass and pH were the responses used in different models to evaluate individua ll y fit ones. Quadratic model was found to be fit for chitinase response whereas in the case of biomass and pH, linear model was found to be fit without th e effect of others. Chitinase production was optimized with respect to other responses such as biomass and pH in multiresponse an alys is of response surface design by using desirability approach. In multiresponse analysis, following medium formulation (gil), chitin, 15 ; urea, 0.32; CaCl 2 , 0.10 and MgS0 4 .7H 2 0, 0.08 was found to predict optimum chitinase production of 482.77 units/ml with overall hi ghest desirability of 0.854 as compared to other formulation s. The selection of model was done on the basis of hi gh Adjusted R-squared va lu e and lo wered p- va lu e for each model in individual analysis of each response. In multiresponse experiment, it was found that for response chi tinase quadratic model and for responses pH and biomass linear models were well fit. Through desirability analysis, it was found that in th e chitinase production, pH was essential as compared to biomass in P. dispersa. Endochitinase and chitobiase actvities were also studied. Keywords: R-Squared value, bi omass, medium components, Pantoea dispersa, pH , p-value, Respon se surface desig n, Chitinase production Experimental design methods have found broad applications in many disciplines. In fact, we may view experimentation as part of the scientific process and as one of the ways of learning how the systems or processes work. Experimental design is a critically important tool in the engineering world for improving the performance of a manufacturing process. It also has extensive applications in the development of new media constituents 1 ' 2 • In a characterization experiment, we are usually interested in determining, which formulation of media variables affect the response. A logical step is to optimize the important factors that leads to the best possible response. There are number of designs available for optimization of process, from these central composite design is one of the most important experimental designs used in the process optimization studies 3 .4. Response surface methodology (RSM) is a collection of mathematical and statistical techniques that are useful for the modeling and analysis of problems in which a response of interest is influenced by several variables *Correspondent author- Phone: +91-265-2794396; Fax: +91-265-2792508 E-mail address: chhatparhs@yahoo.co .i n and objective is to optimize this response 5 . In our present multiresponse experiment, we simultaneously model behavior of the response variables chitinase, biomass and pH in P. dispersa as a function of input variables such as chitin, urea, CaCh and MgS0 4 .7H 2 0 within some region of interest. The model adequacy checking was also carried out, which is necessary to-{1) examine the fitted model to ensure that it provides an adequate approximation to the true system and (2) to verify that none of the response surface model assumptions are violated. Optimization of the multiresponse function using desirability function approach was done, which is an analytical technique based on the concept of utility or desirability associated with a given response function 6 · 7 . Present report was an attempt to formulate a suitable production medium using multiresponse analysis that can substantially increase the chitinase production by P. dispersa. Materials and Methods Organism, culture conditions for the growth-- The culture P. dispersa was isolated in our laboratory from sea dumps, in Bhavnagar, India. The organism was