This study describes the application of response surface methodology in the optimization of guar gum microspheres for colon-specific delivery of metronidazole. The effect of varying the relative percent of the four factors used, that is guar gum, glutaraldehyde, swelling time, and stirring speed, has been systematically investigated for identifying their best values to optimize the drug release and encapsulation efficiency as well as to highlight possible interactions among the components. Different batches were prepared according to 24 factorial designs and randomly evaluated for drug release and drug encapsulation efficiency. Analysis of response surface plots allowed identification of the best combination of four levels to minimize drug release in upper part of gastrointestinal tract and maximize the encapsulation efficiency. The scanning electron microscopy was used to characterize the surface of these microspheres. Drug polymer interactions were assessed by differential scanning calorimetry and XRD. The good correspondence between calculated and experimental values indicated in the validity of the generated statistical model. Only a small fraction of drug was released at acidic pH; however, the release of drug was found to be higher in the presence of rat cecal contents, indicating the susceptibility of crosslinked guar gum matrix to colonic enzymes released from rat cecal contents. Metronidazole release kinetics corresponds best to zero-order model and drug release mechanism was diffusion and swelling controlled. The significance of differences was evaluated by analysis of variance (ANOVA). Differences were considered statistically significant at P < 0.05. (C) 2012 Wiley Periodicals, Inc. J Appl Polym Sci, 2012