Response surface methodology (RSM) was used to optimize microencapsulation yield (MY) using three independent variables; the ratio of coating material to core material (w/w, X1), the emulsifier concentration (%, v/v, X2), and the CaCl2 concentration (%, w/v, X3). In the preparation of sodium alginate (SA) microcapsule, the regression model equation for the MY was predicted as follows; MY(%) = 56.02 + 3.64X2 + 3.18X1X2 - 3.74X2(2). The optimal conditions for the SA microcapsule were obtained at the [SA]/[alpha-TP] ratio of 6.6:3.4 (w/w), [emulsifier] of 1.35% (v/v), and [CaCl2] of 4.3% (w/v), and the predicted MY in this condition was of 57.2%. In vitro alpha-TP releasing test of the SA-based microcapsules was performed. The SA microcapsule released 28.8% of alpha-TP when exposed in the simulated gastric fluid (SGF, pH 1.2) for 24 h. In the simulated intestinal fluid (SIF, pH 7.4), the amount of released alpha-TP (81.5%) was significantly greater than that in the SGF. The duration time required for releasing 50 (T50%) and 70% (T70%) of alpha-TP from the SA-microcapsule were calculated to be 3.8 and 12.3 h, respectively. From these results, it was suggested that SA microcapsule would be structurally resistant against acidic environment, and it would rapidly release core material under mild alkali condition.