Zhi-Sheng Liu

Elimination of trypsin inhibitor activity and beany flavor in soy milk by consecutive blanching and ultrahigh-temperature (UHT) processing.

Soy foods contain significant health-promoting components but also may contain beany flavor and trypsin inhibitor activity (TIA), which can cause pancreatic disease if present at a high level. Thermal processing can inactivate TIA and lipoxygenase. Ultrahigh-temperature (UHT) processing is relatively new for manufacturing soy milk. Simultaneous elimination of TIA and soy odor by UHT processing for enhancing soy milk quality has not been reported. The objective was to determine TIA in soy milk processed by traditional, steam injection, blanching, and UHT methods and to compare the products with commercial soy milk products. Soybean was soaked and blanched at 70-85 degrees C for 30 s-7.5 min. The blanched beans were made into base soy milk. The hexanal content of the base soy milk was determined by gas chromatography to determine the best conditions for further thermal processing by indirect and direct UHT methods at 135-150 degrees C for 10-50 s using the Microthermics processor. Soy milk was also made from soaked soybeans by traditional batch cooking and steaming methods. Eighteen commercial products were selected from the supermarket. Residual TIA in soy milk processed by the traditional and steam injection to 100 degrees C for 20 min was approximately 13%. Blanching could inactivate 25-50% of TIAs of the raw soy milk. The blanch conditions of 80 degrees C and 2 min were selected for UHT processing because these conditions produced blanched soy milk without hexanal, indicating a complete heat inactivation of lipoxygenases. The TIA decreased with increased temperature and time of UHT heating. The maximal trypsin inhibitor inactivation was achieved by UHT direct and indirect methods with residual activities of approximately 10%. Some commercial soy milk products contained high TIAs. The results are important to the food industry and consumers. Kinetic analysis showed that heat inactivation (denaturation) of TIA, under the continuous processing conditions of the Microthermics processor, followed first-order reaction kinetics, and the activation energy of the inactivation was 34 kJ/mol.

Comparative studies on the chemical and cell-based antioxidant activities and antitumor cell proliferation properties of soy milk manufactured by conventional and commercial UHT methods.

The aims of this work were to compare antiproliferation, antioxidant activities and total phytochemicals and individual isoflavone profiles in soy milk processed by various methods including traditional stove cooking, direct steam injection, direct ultrahigh temperature (UHT), indirect UHT, and a two-stage simulated industry method, and a selected commercial soy milk product. Various processing methods significantly affected total saponin, phytic acid, and total phenolic content and individual isoflavone distribution. The laboratory UHT and the two-stage processed soy milk exhibited relatively higher total phenolic content, total flavonoid content, saponin and phytic acid than those processed by the traditional and steam processed methods. Thermal processing caused obvious intertransformation but did not cause severe degradation except for breaking down of aglycons. Thermal processing significantly increased antioxidant capacities of soy milk determined by chemical analyses, but decreased cellular antioxidant capacities as compared to the raw soy milk. The raw and all processed soy milk exhibited antipoliferative activities against human HL-60 leukemia cells, AGS gastric tumor cells, and DU145 prostate cancer cells in a dose-dependent manner. The raw soy milk, but not the processed soy milk, exhibited a dose-dependent antiproliferative effect against colorectal adenocarcinoma Caco-2 cells. Taken together, these results indicate that various thermal processing methods change not only phytochemcials but also potential health-promoting effects of soy milk.

Off-flavor related volatiles in soymilk as affected by soybean variety, grinding, and heat-processing methods.

Off-flavor of soymilk is a barrier to the acceptance of consumers. The objectionable soy odor can be reduced through inhibition of their formation or through removal after being formed. In this study, soymilk was prepared by three grinding methods (ambient, cold, and hot grinding) from two varieties (yellow Prosoy and a black soybean) before undergoing three heating processes: stove cooking, one-phase UHT (ultrahigh temperature), and two-phase UHT process using a Microthermics direct injection processor, which was equipped with a vacuuming step to remove injected water and volatiles. Eight typical soy odor compounds, generated from lipid oxidation, were extracted by a solid-phase microextraction method and analyzed by gas chromatography. The results showed that hot grinding and cold grinding significantly reduced off-flavor as compared with ambient grinding, and hot grinding achieved the best result. The UHT methods, especially the two-phase UHT method, were effective to reduce soy odor. Different odor compounds showed distinct concentration patterns because of different formation mechanisms. The two varieties behaved differently in odor formation during the soymilk-making process. Most odor compounds could be reduced to below the detection limit through a combination of hot grinding and two-phase UHT processing. However, hot grinding gave lower solid and protein recoveries in soymilk.

Chemical characteristics of low-fat soymilk prepared by low-speed centrifugal fractionation of the raw soymilk.

Large oil-protein particles (2 to 60 microm) were found in raw soymilk (or water extract of soybean), which was prepared in specific conditions. The large particles could be separated by sedimentation by centrifuging raw soymilk for 5 to 30 min at a low gravitational force ranging from 96 to 2410 x g. Chemical analysis showed that 80% to 90% of the total lipids and 30% to 40% of the total proteins were located in the precipitated fraction. The supernatant fraction had a dramatically higher protein-to-lipid ratio than the whole soymilk. The ratio of 11S/7S proteins and the ratio of 11S acidic/basic subunits were significantly (P < 0.05) higher in the precipitate than that either in the whole soymilk or in the supernatant. Besides centrifuging conditions, other factors, including soymilk concentration, grinding method, soybean variety, and soybean storage, also significantly (P < 0.05) affected the centrifugal fractionation. This study showed that low-speed centrifugation facilitated the separation of oil-protein particles from raw soymilk, and can be used as an innovative method for preparing low-fat soymilk and 11S protein-enriched ingredients. The findings also increased our understanding of the association or aggregation between proteins and lipids in raw soymilk after grinding.