Factors Affecting Psicose Formation in Food Products during Cooking

Abstract

Factors affecting the formation of psicose in food products during cooking were investigated based on psicose formation from fructose in various food products and during the caramelization process. Psicose was produced upon heating not only in fructose meringue that had a high fructose content and was alkaline, but also in dried apple that was weakly acidic. Optimized conditions for psicose formation were determined by experiments on fructose caramelization. From the results, we concluded that the following factors affect psicose formation in food products during cooking: high temperatures, high pH, high fructose concentration, and extended cooking time. Of these, the pH of the food was the most important factor for psicose formation in food products.

Introduction

d-Psicose (d-ribo-2-hexulose) is a C-3 epimer of uctose that is found in fairly small quantities in nature. Because of the small amount of d-psicose present in natural products, few studies have investigated this sugar. Recently, Izumori and his co-workers developed a new method for the large-scale production of d-psicose (Takeshita et al., 2000). The ability to produce large amounts of d-psicose has made it possible to conduct various studies on its use in food. Safety testing of the sugar has verified that d-psicose causes no mutagenesis or acute toxicity (Matsuo et al., 2002). The sweetness of d-psicose is approximately 70% that of sucrose, and little of the sugar is converted to energy in humans (Iida et al., 2010). Additionally, d-psicose suppresses blood glucose elevation after eating (Hayashi et al., 2010; Iida et al., 2008) and reduces body fat accumulation (Matsuo et al., 2001). Moreover, food products incorporating proteins glycated with d-psicose exhibit excellent antioxidant activity and good rheological properties (Puangmanee et al., 2008; Sun et al., 2006). Hence, the addition of d-psicose as a substitute for alimentary sugars in foodstuffs might be a strategy for developing new functional foods. A syrup containing this sugar is currently on the market.

The commercialization of d-psicose as the functional sugar in a food requires that the factors affecting d-psicose formation in food products be clarified, because foods containing a large amount of d-psicose are expected, so the function of the foods must be accurately estimated. Previously (Oshima et al., 2006) we reported that psicose is found in various food products and suggested that the sugar was produced from fructose or sucrose under ordinary heating conditions. Moreover, based on the psicose concentration in food products, we expected that the sugar concentration, heating time, and temperature during the manufacturing process influence the concentration of psicose in food products. Binkley (1963) reported that d-psicose could be formed from d-fructose when heating limed cane juice at 95 – 97°C for 48 h during the manufacture of molasses. Based on that study, it was suggested that the pH value of the food product also contributes in the extent of psicose formation.

The Lobry de Bruyn-Alberda van Ekenstein rearrangement (Speck, 1958), which is the enolization of sugar anions resulting in monosaccharide isomerisation by non-enzymatic means, is a well-known reaction for the isomerization of sugar. Many investigators (Beveridg et al., 1982; De Bruijn et al., 1987; El Khadem et al., 1987) have studied the formation of d-psicose by the isomerization of d-fructose in aqueous alkaline solutions. However, those studies were inadequate to explain the formation of psicose in food products, because they did not focus on the formation of psicose during cooking.

We hypothesized that the psicose content is closely related to the fructose content in cooked food products. In the present study, to elucidate the mechanism by which psicose is formed in food products during cooking, we investigated the relationship between psicose content and fructose content in dried apple, sponge cake, and fructose meringue as models for different culinary treatment conditions. Additionally, to clarify the factors affecting psicose formation in food products, we carried out a model experiment in cooked food products using the caramelization of fructose at different pH values, temperatures, and fructose concentrations.

Materials and Methods

Chemicals    d-Psicose was purchased from Rare Sugar Production Technical Research Laboratories, LLC (Kagawa, Japan). Unless otherwise specified, all other sugars and chemicals were obtained from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). HPLC-grade water was generated using a Milli-Q water system (Millipore S.A.S., Molsheim, France). Foodstuffs were purchased from local markets in the Takamatsu area, Kagawa, Japan.

Analysis    The sugar content in the food products and in the caramel solution was determined by a post-column HPLC system, which used a gel permeation column in the ligand exchange mode (GL-C611 column [10.7 mm i.d. × 300 mm]; Hitachi, Ltd., Tokyo, Japan) coupled with a pulsed amperometric detector (Oshima et al., 2006). The brown color intensity of the caramel solution was measured using a U-2001 spectrophotometer (Hitachi, Ltd.) at 420 nm at room temperature. The pH value of each sample was obtained using a compact pH-meter (model B-212, Horiba, Ltd., Kyoto, Japan).

Culinary treatment of food products    The dried apple was prepared from sliced apple (5 mm thick) that was set on a tray and heated in a forced convection oven (model FC-612, Toyo Seisakusho Kaisha, Ltd., Chiba, Japan) at 80°C for 24 h. The sponge cake was prepared as follows: egg yolk (60 g) and sucrose (60 g) were mixed using a hand mixer. Egg whites (120 g) and sucrose (30 g) were whipped using a KitchenAid mixer. Both mixtures were mixed together, and wheat flour (90 g) and melted butter (30 g) were added. The cake batter (350 g) was placed in a coated tube pan and heated at 180°C in a bakery oven (model Scorb-4.5 MP, Nichiwa Electric Co., Ltd., Hyogo, Japan) for 0.47 h. The fructose meringue was prepared using the following steps: 30 g of egg white was whipped using an electric mixer and 40 g of fructose was added gradually. Ten grams of the mixture was placed in aluminium cups. The uncovered cups were heated at 100°C in a dry sterilising oven (model SP-650, Toyo Seisakusho Kaisha, Ltd.) for 2 h. The sucrose, glucose, fructose, and psicose content and the pH value of each sample were determined according to the above methods. The sugar content analyses were carried out in triplicate and the mean values were expressed ± the standard deviation.

Caramelization of fructose    The model experiment for the caramelization of fructose was conducted according to a modification of the procedure used by Ajandouz et al. (2001). Fructose solutions (3 mL) of various concentrations [10%, 20%, and 40% (w/w)] were placed in 4-mL screw-cap tubes and heated in a block heater at 60, 80, or 100°C. At specific times (1, 2, 3, 4, 8, 16, and 24 h), each tube was removed from the heater and immediately cooled on ice. After cooling, the fructose, glucose, and psicose content, the brown color intensity, and the final pH of the solutions were analysed using the methods described above. The initial pH adjustment of the reaction mixture was performed using the following buffers: 0.05 M sodium acetate buffer with 1 M acetic acid at pH 4.0, 0.05 M sodium phosphate buffer at pH 6.5, and 0.05 M Tris buffer with 1 M hydrochloric acid at pH 9.0.

Results and Discussion

Formation of psicose during cooking    Previously, we reported that psicose is found in various food products, but the mechanism by which the sugar is produced was not clear (Oshima et al., 2006). In the current study we therefore cooked some model foods using different culinary treatment conditions. Table 1 shows the changes in the sugar composition and pH in dried apple during cooking. Each sugar is expressed as dry weight. Fructose, present at a concentration double that of glucose, was the predominant sugar present in the apple. At time 0, no psicose was detected in the apple, but the sugar increased gradually to 0.11 g/100 g during 24 h of cooking. The content of the other sugars decreased as the concentration of psicose increased: the fructose content was reduced by 20% during cooking. The pH value of the sample did not change during cooking. The ratio of psicose to fructose (psi/fru) was calculated to be 0.003 after cooking for 24 h. These results suggest that psicose is produced under weakly acidic conditions during extended cooking. This is the first report of the formation of psicose in food products under acidic conditions (pH 3.9).

Table 2 shows the changes in the sugar composition and the pH of sponge cake during cooking. Each sugar was expressed as dry weight. Sucrose was the major sugar present in the cake. At time 0, no psicose was detected in the cake. However, during cooking for 0.47 h, psicose increased to 13.0 mg/100 g, while the sucrose content decreased in the cake. The fructose and glucose content doubled and tripled, respectively, during cooking, and were apparently produced by the hydrolysis of sucrose. Cooking did not cause any change in pH. The psi/fru ratio was calculated to be 0.012 after cooking for 0.47 h. These results indicate that cooking for a short time at a high temperature and at a neutral pH leads to the formation of psicose.

Table 1. Changes in the sugar composition and pH in dried apple during cooking

Dried apple
Time (h)	Content (g/100 g)a)				pH	Psi/Fru
	Sucrose	Glucose	Fructose	Psicose
0	7.45 ± 0.27	23.91 ± 1.21	43.04 ± 1.94	ndb)	3.9	−
9	5.88 ± 0.16	21.58 ± 0.99	39.61 ± 1.39	0.02 ± 0.00	3.9	0.000
24	5.24 ± 0.19	20.12 ± 0.97	34.23 ± 1.57	0.11 ± 0.01	3.9	0.003

^a)Expressed as each sugar content/100 g of dry weight. The values are the means of three replicates ± the standard deviations.

^b)Not detected.

Table 2. Changes in the sugar composition and pH in sponge cake during cooking

Sponge cake
Time (h)	Content (g/100 g)a)				pH	Psi/Fru
	Sucrose	Glucose	Fructose	Psicose
0	37466.7 ± 671.0	490.0 ± 14.6	423.3 ± 22.4	ndb)	7.3	−
0.47	34720.0 ± 1197.5	1486.7 ± 72.1	1073.3 ± 82.0	13.0 ± 0.6	7.3	0.012

^a)Expressed as each sugar content/100 g of dry weight. The values are the means of three replicates ± the standard deviations.

^b)Not detected.

Table 3 shows the changes in sugar composition and in the pH of the fructose meringue during cooking. Each sugar was expressed as dry weight. At time 0, glucose and psicose were not detected in the meringue. However, during 2 h of cooking, the glucose and psicose content greatly increased to 3.95 g/100 g and 1.72 g/100 g, respectively. The fructose content was reduced by 9% during 2 h of cooking. The pH of the meringue decreased gradually from 8.5 to 6.7 during this time. Caramelization causes the release of H+ (Kroh, 1994). Thus, the pH drop during cooking may be due to caramelization. The psi/fru ratio was calculated to be 0.020 after 2 h of cooking. These results indicate that psicose formation in the meringue is accelerated under alkaline conditions.

Table 3. Changes in the sugar composition and pH in fructose meringue during cooking

Fructose meringue
Time (h)	Content (g/100 g)a)			pH	Psi/Fru
	Glucose	Fructose	Psicose
0	ndb)	92.04 ± 0.83	ndb)	8.5	−
1	3.07 ± 0.18	85.71 ± 0.89	1.07 ± 0.01	7.5	0.012
2	3.95 ± 0.20	84.21 ± 1.31	1.72 ± 0.05	6.7	0.020

^a)Expressed as each sugar content/100 g of dry weight. The values are the means of three replicates ± the standard deviations.

^b)Not detected.

We previously (Oshima et al., 2006) demonstrated that psicose is produced by sucrose hydrolysis in sucrose caramel sauce. Therefore, we hypothesized that psicose is produced not only from fructose, but also from sucrose via fructose in dried apple and sponge cake. The current results indicate that the amount of psicose formation during cooking differs for different food products. Furthermore, the change in the ratio of fructose to psicose in food products is apparently affected by cooking conditions, since the psi/fru ratio differed for each food product. These results suggest that psicose formation from fructose in cooked food products is affected by pH, temperature, sugar content, and heating time during cooking.

It is well known that caramelization occurs in foods when foods are heated at high temperature. Furthermore, the enolization of sugar occurs as part of the caramelization process (Kroh, 1994). Therefore, to clarify the factors affecting psicose formation in food products during cooking, we carried out a model experiment on the caramelization of fructose.

Effect of pH on psicose formation    The effect of pH on psicose formation in the reaction mixture was examined at 100°C for 24 h with a 40% (w/w) aqueous solution of fructose. The initial pH of the reaction mixture was adjusted to 4.0, 6.5, and 9.0, which are typical pH values for food during cooking (Fig. 1). A higher initial pH of the reaction mixture led to an increase in psicose content and a decrease in fructose content. The psicose content in the reaction mixtures adjusted to pH 6.5 and 9.0 increased rapidly in the early stages of the process, then became constant after 4 h. In contrast, the psicose content increased slightly for 24 h for the pH 4 reaction mixture (Fig. 1A). The glucose content of the reaction mixtures with initial pH values of 6.5 and 9.0 rapidly increased in the early stages of the process, then became constant after 8 h and 4 h, respectively. At an initial pH of 4.0, the glucose content of the reaction mixture increased gradually throughout the process (Fig. 1B). The ratio of psicose to glucose (psi/glc) in the reaction mixture with initial pH values of 6.5 and 9.0 became almost constant (between 0.42 and 0.55) after 1 h. In contrast, at an initial pH of 4.0, the ratio gradually increased to 0.2 over 24 h. These results suggest that the initial pH of the reaction mixture affects the equilibrium between psicose and glucose. At initial pH values of 6.5 and 9.0, the fructose content rapidly decreased during the early stages of the process, then decreased slowly after 8 h (Fig. 1C). The psi/fru ratio was calculated to be 0.002, 0.014, and 0.036 at initial pH values of 4.0, 6.5, and 9.0, respectively, after 24 h. Thus, reaction conditions providing a higher initial pH led to an increase in the psi/fru ratio. For the samples with an initial pH of 6.5 or 9.0, the final pH in the reaction mixture dropped during the early stages and stabilized at around pH 4. On the other hand, samples with an initial pH of 4.0 did not change in pH during the reaction (Fig. 1D). De Bruijn et al. (1987) reported that the observed pH drop was due to an increase in the acid concentration during caramelization. Therefore, acid produced in the pH 6.5 and 9.0 reaction mixtures likely caused the pH value of the mixtures to drop.

Fig. 1.

The effect of pH on psicose formation in a 40% (w/w) aqueous solution of fructose adjusted to pH 4.0, 6.5, or 9.0 and heated at 100°C for 24 h. The psicose, glucose and fructose contents are shown as the amount per 100 g of reaction mixture. A, Formation of psicose; B, Formation of glucose; C, Reduction in fructose; D, Final pH; E, Browning.

Raising the pH of the reaction mixture led to an increase in the brown color during the early stages of the process. In each heated solution, the brown color increased in intensity approximately linearly, even after the final pH values stabilized at around pH 4 and psicose formation was controlled. In addition, the brown color intensity of samples with initial pH values of 4.0, 6.5, and 9.0 was almost equal after 24 h (Fig. 1E).

Effect of temperature on psicose formation    The effect of temperature on psicose formation in the reaction mixture was examined at 60°C, 80°C, and 100°C for 24 h using 40% (w/w) aqueous solutions of fructose. The initial pH of the reaction mixture was adjusted to 6.5, the pH value of many foods (Fig. 2). The amount of psicose increased with increased reaction temperature and the amount of fructose decreased. The psicose content of the reaction mixture at 100°C increased rapidly during the early stages of the process and then became constant after 4 h. In the reaction mixture at 80°C, psicose gradually increased during the reaction (Fig. 2A). At 100°C, the glucose content was about double that of the psicose content, although very little psicose and glucose were detected at 60°C (Fig. 2B). Distinct values for the reaction mixtures at 60°C and 80°C were not observed, although the psi/glc ratio in the 100°C mixture essentially stabilized at 0.45 after 8 h. Under these reaction conditions, the fructose content of the reaction mixtures at 60°C and 80°C decreased more slowly than at 100°C (Fig. 2C). After 24 h, the psi/fru ratio in the reaction mixture was 0.001 at 60°C, 0.010 at 80°C, and 0.014 at 100°C. Clearly, raising the reaction temperature led to an increase in the psi/fru ratio. At 100°C, the final pH stabilized at approximately pH 4, but at 80°C the pH decreased to pH5 after 24 h, whereas the pH did not change substantially during incubation at 60°C (Fig. 2D).

Fig. 2.

The effect of temperature on psicose formation in a 40% (w/w) aqueous solution of fructose adjusted to pH 6.5 and heated at 60°C, 80°C, or 100°C for 24 h. The psicose, glucose and fructose contents are shown as the amount per 100 g of reaction mixture. A, Formation of psicose; B, Formation of glucose; C, Reduction in fructose; D, Final pH; E, Browning.

At 100°C, the brown color intensity increased approximately linearly, even after the final pH value stabilized at around pH 4. The intensity at 80°C was approximately 1/10 that of the intensity at 100°C, while no change in color occurred at 60°C (Fig. 2E).

Effect of fructose concentration on psicose formation    The effect of the fructose concentration in the reaction mixture on psicose formation was examined at 100°C for 24 h using solutions of different concentrations of fructose (10%, 20%, and 40% [w/w]). The initial pH of the reaction mixtures was 6.5, similar to the pH value of many foods (Fig. 3). The higher the initial fructose concentration in the reaction mixture, the higher the final concentration of psicose. In all cases, the psicose content rapidly increased during the early stages of the process and then stabilized after 8 h for the 10% fructose solution and after 4 h for the 20% and 40% fructose solutions (Fig. 3A). A higher initial fructose concentration in the reaction mixture resulted in an increased glucose concentration (Fig. 3B). The psi/glc ratio in the reaction mixture decreased with increased initial fructose concentration, and stabilized at around 0.60, 0.50, and 0.45 in 10%, 20%, and 40% fructose solution, respectively. In reaction mixtures of each fructose concentration tested, the fructose content gradually decreased during heating (Fig. 3C). The psi/fru ratio in each reaction mixture increased at similar rates up to 4 h, then the ratio decreased with increasing initial fructose concentration. After 24 h, the ratio was calculated to be 0.038 in 10% fructose solution, 0.024 in 20% fructose solution, and 0.014 in 40% fructose solution. The final pH of each reaction mixture gradually decreased over time; the extent of this decrease was higher at higher initial fructose concentration (Fig. 3D). This result was attributed to the fact that acid production increased (De Bruijn et al., 1987) with increasing initial fructose concentration in the reaction mixture. The brown color intensity of the reaction mixture increased linearly with increasing initial fructose concentration. The color intensity of 10% fructose solutions was approximately 1/3 the intensity of 40% fructose solution, and the color intensity of 20% fructose solution was approximately half the intensity of 40% fructose solution (Fig. 3E).

Fig. 3.

The effect of fructose concentration on psicose formation in different concentrations of fructose solutions (10%, 20%, and 40% [w/w]) adjusted to pH 6.5 and heated at 100°C for 24 h. The psicose, glucose and fructose contents are shown as the amount per 100 g of reaction mixture. A, Formation of psicose; B, Formation of glucose; C, Reduction in fructose; D, Final pH; E, Browning.

Factors affecting psicose formation during cooking    In experiments in which psicose was formed during cooking, psicose was produced from fructose in cooked food products, and the psi/fru ratio in cooked food products was affected by the cooking condition (Table 1 – 3). These results were corroborated by experiments on fructose caramelization. For example, in dried apple, psicose was produced by heating for a long time despite the acidic conditions, but the psi/fru ratio was small (0.003; Table 1). In the fructose caramelization experiment, psicose was produced from fructose both under alkaline (pH 9.0) and acidic (pH 4.0) conditions. However, the psi/fru ratio under acidic conditions after incubation for 24 h was small (0.002; Fig. 1A). For neutral and alkaline foods, psicose formation was facilitated by heating for a short time, and the psi/fru ratio was large. This result was observed for sponge cake, meringue, and for caramelization (Tables 2, 3, and Fig. 1A).

Moreover, psicose formation was strongly influenced by the heating temperature (Fig. 2). The sponge cake was heated at a higher temperature than the other tested foods, so this result suggested that psicose content was increased by heating for a short time (Table 2). Furthermore, psicose formation was high in food products with a high fructose content, as in the case of fructose meringue and caramelization (Table 3 and Fig. 3A). In addition, during caramelization, the psicose content increased with increased heating time, although the rate of psicose formation was inhibited by a pH drop under these reaction conditions (Fig. 1–3A and D). In fructose meringue, similar tendencies were observed: the psicose formation rate after 1 to 2 h heating decreased relative to the rate observed during the first hour of heating (Table 3). In the fructose caramelization experiment, the psi/fru ratio fluctuated according to changes in the cooking conditions: i.e., temperature, pH, sugar concentration, and time. Moreover, in the experiments studying psicose formation in various food products (dried apple, sponge cake, and fructose meringue) these cooked food products were shown to have a characteristic psi/fru ratio depending on the cooking conditions.

On the other hand, in the fructose caramelization experiment, the brown color intensity increased with psicose formation during the early stage of the process and the intensity increased linearly over time, even though the rate of psicose formation decreased due to a pH drop. These results suggest a rough correlation between psicose content and brown color intensity in the caramelization experiment (Fig. 1–3A and C). Development of a brown color was confirmed in dried apple, sponge cake, and fructose meringue in which psicose was formed during cooking (data not shown). Therefore, it is expected that food products that turn brownish during thermal processing contain a large amount of psicose. Food products contain various ingredients (e.g., amino acids, protein, polysaccharide, and inorganic salts) and these ingredients were assumed to influence psicose formation. However, as the results of the experiments with fructose caramelization indicate, psicose formation might be easily caused by browning during cooking at high temperatures, by high pH, by high fructose concentration, and by a longer cooking time. In particular, the pH appears to be the most important factor for psicose formation.

The factors affecting psicose formation presented in this paper are in accordance with our previous paper (Oshima et al., 2006). Worcester sauce had the highest psicose content among the studied food products. Fruits and sugars are common ingredients in Worcester sauce, and they are simmered over a long period during Worcester sauce manufacturing. Because of this long manufacturing process, the fructose in the fruits and sugar might be converted to psicose under acidic conditions. A high psicose content was confirmed in food products that were heated at high temperatures, such as tinsukou (Japanese cookie), kawarasenbei (Japanese rice cracker), and corn-snacks, and in food products exposed to alkaline conditions during the manufacturing process, such as brown sugar. Food products colored by heating, mentioned above, contained a large amount of psicose.

The results obtained from our study may be generally applicable to food products, as demonstrated in our previous paper: that is, psicose formation in food products may be controlled by cooking conditions. Examination of the psi/fru ratio in food products might allow estimation of the heating history (e.g., cooking and/or sterilization temperature) during the manufacturing process, since this ratio has a characteristic value depending on the cooking conditions.

Conclusion

The factors affecting psicose formation in cooked food products were investigated in this study. The results suggested that in cooked food products, psicose formation increases with the following cooking conditions: temperature, pH, fructose concentration, and time. In particular, the pH value of the foods was the most important factor for the production of psicose. In addition, it is likely that food products turned brownish by thermal processing contain psicose. A syrup (currently marketed) containing psicose was developed on the basis of this research. In addition, the psi/fru ratio in food might function as a useful tool for investigating their heat histories.

Acknowledgements    This work was supported in part by a grant from the Technical Development Association Project (Glyco-Bio- Cluster) of the Kagawa Prefectural Government, Japan. We wish to thank Mrs. Noriko Kagawa, Ms. Emiko Yokota, and Ms. Yuka Ozaki, Food Research Branch, Kagawa Industrial Technology Centre, for their helpful collaboration.

References

•  Ajandouz,  E. H.,  Tchiakpe,  L. S.,  Ore  F. D.,  Benajiba,  A., and  Puigserver,  A. (2001). Effects of pH on caramelization and maillard reaction kinetics in fructose-lysin model systems. J. Food Sci., 66, 926-931.
•  Beveridg,  R. J.,  Davis,  M., and  Morris,  J. L. (1982). The preparation of decagram quantities of d-psicose by the isomerization of d-fructose, and separation of the products on a calcium-ion cation-exchange resin. Carbohydrate Research, 101, 348-349.
•  Binkley,  W. W. (1963). The fate of can juice simple sugars during molasses formation IV. Probable conversion of d-fructose to d-psicose. Int. Sugar J., 65, 105-106.
•  De Bruijn,  J. M.,  Kieboom,  A. P. G., and  van Bekkum,  H. (1987). Alkaline degradation of monosaccharides V: Kinetics of the alkaline isomerization and degradation of monosaccharides. Recl. Teav. Chim. Pays-Bars, 106, 35-43.
•  El Khadem,  H. S.,  Ennifar,  S., and  Isbell,  H. S. (1987). Contribution of the reaction pathways involved in the isomerization of monosaccharides by alkali. Carbohydrate Research, 169, 13-21.
•  Hayashi,  N.,  Iida,  T.,  Yamada,  T.,  Okuma,  K.,  Takehara,  I.,  Yamamoto,  T.,  Yamada,  K., and  Tokuda,  M. (2010). Study on the postprandial blood glucose suppression effect of d-psicose in borderline diabetes and the safety of long-term ingestion by normal human subjects. Biosci. Biotechnol. Biochem., 74, 510-519.
•  Iida,  T.,  Hayashi,  N.,  Yamada,  T.,  Yoshikawa,  Y.,  Miyazato,  S.,  Kishimoto,  Y.,  Okuma,  K.,  Tokuda,  M., and  Izumori,  K. (2010). Failure of d-psicose absorbed in the small intestine to metabolize into energy and its low large intestinal fermentability in humans. Metabolism, 59, 206-214.
•  Iida,  T.,  Kishimoto,  Y.,  Yoshikawa,  Y.,  Hayashi,  N.,  Okuma,  K.,  Tohi,  M.,  Yagi,  K.,  Matsuo,  T., and  Izumori,  K. (2008). Acute d-psicose administration decreases the glycemic responses to an oral maltodextrin tolerance test in normal adults. J. Nutr. Sci. Vitaminol., 54, 511-514.
•  Kroh,  L. (1994). Caramelisation in food and beverages. Food Chem., 51, 373-379.
•  Matsuo,  T.,  Baba,  Y.,  Hashiguchi,  M.,  Takeshita,  K.,  Izumori,  K., and  Suzuki,  H. (2001). Less body fat accumulation with D-psicose diet versus D-fructose diet. J. Clin. Biochem. Nutr., 30, 55-65.
•  Matsuo,  T.,  Tanaka,  T.,  Hashiguchi,  M.,  Izumori,  K., and  Suzuki,  H. (2002). Effects of oral acute administration and subchronic feeding of several levels of d-psicose in rats. J. Nutr. Sci. Vitaminol., 48, 512-516.
•  Oshima,  H.,  Kimura,  I., and  Izumori,  K. (2006). Psicose contents in various food products and its origin. Food Sci. Technol. Res., 12, 137-143.
•  Puangmanee,  S.,  Hayakawa,  S.,  Sun,  Y., and  Ogawa,  M. (2008). Application of whey protein isolate glycated with rare sugars to ice cream. Food Sci. Technol. Res., 14, 457-466.
•  Speck, Jr. J. C. (1958). The Lobry de Bruyn-Alberda van Ekenstein transformation. Ad. Carbohydr. Chem., 13, 63-103.
•  Sun,  Y.,  Hayakawa,  S.,  Jiang,  H.,  Ogawa,  M., and  Izumori,  K. (2006). Rheological characteristics of heat-induced custard pudding gels with high antioxidative activity. Biosci. Biotechnol. Biochem., 70, 2859-2867.
•  Takeshita,  K.,  Suga,  A.,  Takada,  G., and  Izumori,  K. (2000). Mass production of d-psicose from d-fructose by continuous bioreactor system using immobilized d-tagatose 3-epimerase. J. Biosci. Bioeng., 90, 453-455.