Horseradish peroxidase (HRP) is an important heme-containing glyco-enzyme that has been used in many biotechnological fields. Valuable proteins like HRP can be obtained in sufficient amounts using Escherichia coli as an expression system. However, frequently, the expression of recombinant enzyme results in inclusion bodies, and the refolding yield is generally low for proteins such as plant peroxidases. In this study, a recombinant HRP was cloned and expressed in the form of inclusion bodies. Initially, the influence of few additives on HRP refolding was assessed by the one factor at a time method. Subsequently, factors with significant effects including glycerol, GSSG/DTT, and the enzyme concentration were selected for further optimization by means of the central composite design of response surface methodology (RSM). Under the obtained optimal condition, refolding increased about twofold. The refolding process was then monitored by the intrinsic fluorescence intensity under optimal conditions (0.35 mM GSSG, 0.044 mM DTT, 7 % glycerol, 1.7 M urea, and 2 mM CaCl2 in 20 mM Tris, pH 8.5) and the reconstitution of heme to the refolded peroxidase was detected by the Soret absorbance. Additionally, samples under unfolding and refolding conditions were analyzed by Zetasizer to determine size distribution in different media.