Original Communication | Published:

Effects of single and long-term administration of wheat albumin on blood glucose control: randomized controlled clinical trials



To examine the effects of single and long-term administration of wheat albumin (WA) on blood glucose levels and blood glucose control, respectively.


Randomly arranged crossover trial for single administration in healthy subjects and double-blinded randomized controlled trial for long-term administration (3 months) in diabetic patients. In vitro α-amylase inhibitory activity of WA was also determined.


Central Research Laboratories of Nisshin Flour Milling Co. Ltd. (Saitama, Japan) for single administration and Aiwa Clinic (Saitama, Japan) for long-term administration.


A total of 12 healthy adult male volunteers for the single administration and 24 type II outpatient diabetics with mild hyperglycemia for the long-term administration.


Subjects took soups containing 0, 0.25, 0.5, and 1.0 g WA before test meals for single administration, and patients took soups with or without 0.5 g WA before every meal for the long-term (3 months) administration.


In vitro α-amylase inhibitory activity of WA was 150 times that of wheat flour. In the single administration experiment, WA suppressed peak postprandial blood glucose levels in a dose-dependent manner: 31, 47, and 50% reduction after 0.25, 0.5, and 1.0 g administrations, respectively. In the long-term administration, 0.5 g of WA did not affect fasting blood glucose levels, whereas it reduced hemoglobin A1c levels. No significant adverse effects such as hypoglycemia or gastrointestinal disturbances were observed in the two experiments.


In the treatment of type II diabetic patients, WA might be a useful functional food, which, with diet and exercise, could help to improve blood glucose control without any critical adverse effects.


Diabetes mellitus (DM) is one of the major health problems in developed and developing countries. The Japanese Ministry of Health and Public Welfare (1999) conducted a survey in 1997, which estimated that 13.7 million people suffered from impaired glucose tolerance (IGT) or DM, including 6.9 million DM patients in Japan. The latest survey conducted in 2002 revealed that the prevalence grew to reach 16.2 million (10.6% of total population) Japanese people suffering from IGT or DM (Japanese Ministry of Health, Labor, and Public Welfare, 2003). Of the Japanese DM, 90% is classified as type II or noninsulin-dependent DM (NIDDM). The number of patients with type II diabetes has been increasing throughout the world, with World Health Organization (WHO) estimates being 157.3 million in 2000 and 215.6 million in 2010. Asia is not immune from this increase with incidence figures expected to increase from 86.6 million in 2000 to 126.2 million in 2010 (Scherbaum & Zimmet, 1995). Therefore, the treatment and prevention of type II diabetes are important worldwide issues.

The primary goal of treatment of type II diabetes is to maintain blood glucose levels within the physiological range. For those with mild hyperglycemia, this is achieved primarily by diet and exercise; however, this is often difficult to achieve for a long time period. The practical target has therefore been to achieve adequate blood glucose control to reduce the risk of diabetic complications such as retinopathy, nephropathy and neuropathy or atherosclerosis. Large-scale trials have confirmed that glycemic control can prevent the progression of these complications (The Diabetes Control and Complications Trial Research Group, 1993; Stratton et al, 2001), with the Kumamoto study showing no worsening of retinopathy and neuropathy during 6 y of follow-up when the hemoglobin A1c (HbA1c) levels were maintained below 6.5% (Ohkubo et al, 1995).

Several pharmacological agents, for example, sulfonylureas, nonsulfonylurea secretagogues, biguanides, insulin sensitizers, and recently insulin have been used to treat type II diabetes; however, some of these agents have the potential to increase the risk of the adverse effect of severe hypoglycemia (UKPDS, 1998; Saloranta et al, 2002). Inhibitors of carbohydrate digestion and absorption such as α-glucosidase inhibitors have been reported to improve blood glucose control with a low risk of hypoglycemia (Josse et al, 2003; Baba, 1994).

Wheat α-amylase inhibitor is another food material attracting attention because it delays carbohydrate digestion and absorption by a different mechanism to that of an α-glucosidase inhibitor. The amylase inhibitor inhibits conversion of polysaccharides to disaccharides, while the α-glucosidase inhibitor inhibits the conversion of disaccharides to monosaccharides. In 1943, Kneen and Sandstedt discovered α-amylase inhibiting activity in wheat kernel, and in 1976, DePonte et al determined the specific proteins existing in the WA fraction which were responsible for the activity. The specific α-amylase inhibiting proteins were identified as 0.19-, 0.28-, 0.36-, and 0.53-inhibitors according to the results of gel electrophoresis (Sodini et al, 1970). The primary and tertiary structures of these inhibitors were then clarified (Oda et al, 1997; Maeda et al, 1983; Kashlan & Richardson, 1981). Koike et al (1995) found that WA inhibited the increase of postprandial hyperglycemia, and 0.19-inhibitor (0.19-albumin) was revealed to exert the main α-amylase inhibiting activity in WA (Choudhury et al, 1996).

The purposes of the present study were to examine the effects of single and long-term administration of WA on blood glucose levels in normal subjects and blood glucose control in mildly diabetic patients, respectively, by randomly arranged clinical trials. We also determined the α-amylase inhibiting activity of WA by comparing it with that of wheat flour or purified 0.19-albumin.

Materials and methods

WA preparation and in vitro assay of α-amylase inhibiting activity

WA was prepared by the method previously described by Choudhury et al (1996). The content of 0.l9-albumin in WA was 24.9% (w/w). The pure 0.19-albumin was purified by high-performance liquid chromatography (HPLC).

Inhibiting activities of 0.19-albumin and WA against human amylases were assayed using soluble starch as substrate and measuring the residual starch by coloration with iodine. Briefly, the amylase and each test sample were preincubated at 37°C for 30 min in the assay buffer (40 mM PIPES (pH 6.9), 100 mM sodium chloride, 10 mM calcium chloride, and 0.02% ovalbumin), and then the soluble starch in the same buffer was added so as to adjust its concentration to 1.5% w/v. This mixture was incubated for a further 10 min. The α-amylase activity was calculated by coloration of the residual starch with iodine. An inhibiting unit (IU) was defined as the amount required to inhibit the α-amylase activity by 50%, and expressed as IU/mg. Human pancreas and salivary α-amylase were obtained from Athens Research and Technology (Athens, GA, USA) and Sigma (St Louis, MO, USA), respectively.

Effects of single administration of WA in normal subjects


Powdered consommé soup containing 0 (placebo), 0.25,. 0.5, and 1.0 g of WA were used. The 0.19-albumin was confirmed to remain stable even after powdered soup was dissolved in hot water.


In all, 12 healthy adult male volunteers, ranging from 30 to 60 y of age, were recruited in this experiment. Exclusion criteria were extreme obesity or thinness, current illness, a past history of gastrectomy or intestinal tract resection, a known grain allergy, or other possibly critical diseases. Informed consent was obtained from every subject after explaining the objectives, methods, expected effect, and possible adverse effects of the study. All the procedures were conducted under ethical standards based on the Declaration of Helsinki.

Of the 12 subjects enrolled in the experiment, the data from one subject were excluded as the blood glucose levels did not increase after taking the test meal. In the remaining 11 subjects, the biomedical parameters were as follows: age, 41±0.7 y (range: 31–51 y); height, l7l±0.8 cm; body weight, 69±0.8 kg; body mass index (BMI), 23.7±0.16 kg/m2; and fasting blood glucose (FBG) level, 95±0.7 mg/dl (mean±standard error (s.e.)).

Study design

Each subject received four tests using different doses of WA (0, 0.25, 0.5, and 1.0 g) with at least 3 days' interval between any of the two tests. A crossover study was conducted in a single-blind manner, in which the physician randomly assigned the order of the samples for administration without the subjects noticing them.

Each subject was instructed to fast from 21:00 the previous night except for taking water on the morning of the test. On the day of the test, each subject visited the institute and took the soup dissolved in 100 ml of hot water, with a 560 kcal test meal (220 g of rice, 40 g of roast ham, 50 g of boiled egg, 15 g of mayonnaise, one tomato, and 200 g of unsweetened tea) as breakfast at 09:00 o'clock. Immediately prior to and 30 min, 1, 2, and 3 h after the meal, blood samples were collected for measurement of blood glucose and serum insulin levels. These samples were stored frozen (−20°C) until assayed. Smoking was not allowed during the experiments. Subjects were monitored for symptoms of such as nausea, vomiting, diarrhea, loose bowel motions, borborygmi, flatulence, abdominal pain, and constipation throughout the experiments. In addition to direct measurements, the area under the curve (AUC) for postprandial blood glucose and serum insulin levels was calculated. An insulinogenic index was also calculated by dividing the increase in blood glucose by the increase in serum insulin over the first 30 min after the meal.

Effects of long-term administration of WA in mildly diabetic patients


In the experiment, consommé soup powders weighing 5.5 g and containing 0.5 g of WA were used as a test soup. A similar consommé powder without WA (5.0 g/serving size) was used as a placebo soup. Each test and placebo soup was sealed in a similar aluminum film package. Test soups were analyzed by HPLC to confirm that the 0.19-inhibitor content was in the range from 125 to 150 mg per serving (mean, 135 mg).

Subjects and study protocol

The type II diabetic patients recruited for this study were outpatients with mild hyperglycemia, ranging in age from 35 to 70 y (55.9±2.23, mean±s.e.). All subjects were treated with dietary and exercise therapy, and some of them with oral hypoglycemic agents (sulfonylureas); participants using oral hypoglycemic agents at the start of this study were maintained on the same therapy throughout the study without any change of dosage. They had a BMI in the range of 20–28 kg/m2 (24.1±0.67, mean±s.e.), HbA1c of 6.0–8.0% (6.8±0.11, mean±s.e.), and a variation of fasting blood glucose values no more than 30 mg/dl prior to the start of the experiment. Exclusion criteria were patients using insulin, those with serious complications, those with gastrointestinal diseases, those with a history of gastric or intestinal resection, those who were pregnant, those with other critical diseases, or those with an allergy to grain products. Other patients considered ineligible based on the attending physician's evaluations were also excluded.

Written informed consent was obtained from all participants prior to entry in the study. The information provided included the objectives, methods, expected effects, and possible adverse effects of the study. Patients were also informed that they were not compelled to enter the study, and that they could withdraw from the study at any time even after giving consent. All the procedures were conducted under ethical standards based on the Declaration of Helsinki.

In all, 24 patients were divided randomly into the WA group (n=18) and placebo group (n=6) in a double-blind manner. Each patient took a given soup dissolved in approximately 100 ml of hot water at the beginning of each of the three daily meals. They visited the hospital prior to and at 1, 2, and 3 months during the experiment. Body weight was measured and blood and urine samples were collected for monitoring biochemical parameters at each visit. The parameters assessed were FBG, hemoglobin A1c (HbA1c), total cholesterol, HDL-cholesterol, triglycerides, total protein (TP), albumin (ALB), alkaline phosphatase (ALP), lactate dehydrogenase (LDH), glutamic–oxaloacetic transaminase (GOT), glutamic–pyruvic transaminase (GPT), total bilirubin, blood urea nitrogen (BUN), serum creatinine, and urinalysis (glucose and protein qualitatively).

Each patient was instructed to record each meal intake and to check for gastrointestinal symptoms; for example, abdominal pain, vomiting, diarrhea, and so on. At each visit, the physician assessed the patient's condition by evaluating the intake record and assessing the checklist for possible adverse effects.


Plasma glucose was measured by the glucose oxidase method, and serum insulin was measured by the double antibody method using the Human Insulin EIA kit (Yanaihara Institute, Japan). HbA1c was assessed by the Latex–agglutination method. Other biochemical parameters in the blood were determined by Hitachi 7170 autoanalyzer. Urinalyses were conducted using the diagnostic paper UriAce (Terumo, Japan).

Statistical analysis

All numerical values were expressed as means±s.e. For the single administration experiment, comparisons between groups were performed by two-way ANOVA followed by Tukey's honestly significant difference test for pairwise comparison and Dunnett's test for comparison with control group. And regarding the long-term experiment, Student's t-test was used to compare FBG and HbA1c between the control group and WA group, and changes in the FBG and HbA1c in WA group were determined by two-way ANOVA followed by Dunnett's test. The calculations were carried out using Microsoft Excel 2000 with an add-in software Statcel2 (Yanai, 2004). Differences were considered significant at P<0.05.


α-Amylase inhibiting activity in vitro

α-Amylase inhibiting activities of wheat flour, WA, and 0.19-albumin against human amylases are summarized in Table 1. The inhibiting activities of WA, as compared with that of wheat flour, against salivary and pancreatic amylases were 147- (7770 vs 53) and 145- (7840 vs 54) fold, respectively. The inhibiting activity of 0.19-albumin was as much as 374- (19800 vs 53, against salivary amylase) and 376-(20300 vs 54, against pancreatic amylase) fold that of wheat flour, this being at least three times the inhibiting activity of wheat albumin as compared to wheat flour against amylases from salivary glands and pancreas, respectively.

Table 1 Inhibitory activity of wheat flour, WA, and 0.l9-albumin against human salivary and pancreatic amylases

Effects of single administration of WA in normal subjects

The blood glucose kinetics are summarized in Figure 1. After consuming the placebo soup and the test meal, mean blood glucose levels peaked at 30 min, and then returned to the premeal level. The peak blood glucose levels following WA-containing soup were reduced in a dose-dependent manner. Compared to the peak levels of the placebo group, those of the WA groups were 31% (not significant), 47% (P<0.05), and 50% (P<0.05) lower in the 0.25, 0.5, and 1.0 g WA-taking groups, respectively.

Figure 1

Effects of WA-containing soup on postprandial blood glucose concentration (mean). Asterisk indicates the statistical significant difference seen (P<0.05; Tukey's test).

As shown in Figure 2, the AUC of blood glucose levels up to 1 h after the WA intake were 29% (0.25 g WA), 53% (0.5 g WA), and 64% (1.0 g WA) lower than those of the placebo group. Significant differences were observed between the groups of placebo soup and soups containing 0.5 or 1.0 g of WA (both P<0.01).

Figure 2

Effects of WA-containing soup on AUC of blood glucose levels up to 1 h after the intake. Bars and error bars indicate mean and s.e., respectively. Asterisk indicates the statistical significant difference observed (P<0.01; Dunnett's test).

Mean serum insulin levels 30 min after taking WA-containing soups for the three different dosages tended to be lower than those following ingestion of the placebo soup, but there were no significant differences among the four groups at any of the time points (Figure 3). No difference in insulinogenic index was observed between the WA and placebo groups.

Figure 3

Effects of WA-containing soup on postprandial serum insulin level (mean).

Throughout the experimental period, none of the subjects had hypoglycemia or symptoms of gastrointestinal disturbances such as diarrhea, increased flatulence, abdominal pain, vomiting, nausea, and so on. Biochemical parameters in the blood were also not affected by the WA treatments (data not shown).

Effects of long-term administration of WA in mildly diabetic patients

One patient in the WA group dropped out from the study at day 9. This patient was taking an oral hypoglycemic drug and reported hypoglycemic symptoms, which were judged to be related to the hypoglycemic sulfonylurea drug. After the 3-month experimental period, all the cases were assessed whether or not each patient satisfied the inclusion and exclusion criteria before opening the key by the members of the committee, and 18 cases (five cases in placebo group and 13 cases in the WA group) were adopted for further analyses. The clinical characteristics of patients involved in this study are shown in Table 2. No differences in age, sex, and duration of diabetes were observed between the WA and placebo groups.

Table 2 Clinical characteristics of participants in the study involving the long-term administration of WA

In the WA-treated group, FBG and HbA1c levels tended to decline, such that the HbA1c levels at 2 and 3 months were significantly lower than the initial level (Figure 4). Moreover, when the WA group was subdivided into two subgroups (initial HbA1c 6.0–6.9% and 7.0–8.0%), the declining tendency was more apparent in the higher initial HbA1c group, because the levels both at 2 and 3 months were consistently lower than the initial level (Figure 5).

Figure 4

Effects of long-term administration of WA on fasting blood glucose and serum HbA1c levels in type II diabetes patients. Symbols are as follows: , mean FBS in placebo group; ♦, mean FBS in WA group; , mean HbA1c in placebo group; and •, mean HbA1c in WA group. Error bars indicate s.e. Asterisk indicates the statistical significant difference (P<0.05; Dunnett's test).

Figure 5

Effects of long-term administration of WA on serum HbA1c levels in type II diabetic patients with high and low HbA1c levels. Initial HbA1c levels were stratified as follows: 6 to <7 low; and 7 to ≤8 high. Asterisk indicates the statistical significant difference observed (P<0.05; Dunnett's test).

Treatment in the placebo and WA groups was not associated with hypoglycemia or any gastrointestinal symptoms: diarrhea, loose bowel movement, borborygmi, flatulence, abdominal pain, hypophagia, nausea, vomiting, constipation, hyperphagia, and stomach indisposition. One patient in the WA group reported temporary constipation and some increased flatulence after 2 months; however, the symptoms were mild and did not disturb the continuation of treatment. One case in placebo group did not visit at 2 months; therefore, the biochemical data of the patient were lacking at 2 months period (Table 3). The body weight, total cholesterol, HDL-cholesterol, triglyceride, TP, ALB, ALP, LDH, GOT, GPT, total bilirubin, BUN, and creatinine levels did not change in either of the treatment groups (Table 3). Urinalyses were not different between the two groups (data not shown).

Table 3 Effect of long-term ingestion of WA on biochemical parameters in patients with type II diabetes mellitus


In the in vitro study, we confirmed that relative α-amylase inhibiting activity in WA was approximately 150-fold, while that in 0.19-albumin was approximately 375-fold that of wheat flour. Although 0.19-albumin consisted of 24.9% of total WA, it contributed approximately 80% of the inhibiting activities against α-amylases both from the salivary glands and pancreas. In accordance with the results of Choudhury et al (1996), our results confirmed that WA had a strong α-amylase inhibiting activity and its main part was contributed by 0.19-albumin.

In the single administration experiment using WA, the carbohydrate provided approximately 60% of the energy content of the test meal, and thus, postprandial blood glucose was well elevated by the loading (Figure 1). The administration of 0.5 and 1.0 g of WA significantly suppressed the peak levels of postprandial blood glucose, whereas at 3 h after administration, although statistically not significant and the levels were within the normal range, mean blood glucose levels of the WA groups were higher than that without WA (Figure 1). Since WA is a mixture of proteinous components, it could be digested by proteinases over time, and rapidly lost its amylase-inhibiting activity. This would account for the higher blood glucose levels at 3 h in the WA groups. Therefore, the action of WA could be considered to be that of delaying the absorption of carbohydrate, but not that of reducing the amount of carbohydrate absorption.

We presume that the results of suppression of postprandial increase in blood glucose without change of insulin secretion might give β-cells rest through the insulin-sparing effect of wheat albumin. This may be beneficial to prevent fatigue of β-cells.

The results of the single administration trial confirmed the promising benefit of WA on blood glucose control without significant adverse effects in normal subjects. Therefore, the efficacy was further evaluated by the long-term clinical trials in mildly diabetic patients.

These showed that WA administration failed to reduce FBG levels but reduced HbA1c levels in subjects after 2 and 3 months of treatment (Figure 4), and the effect was more apparent in subjects with higher initial HbA1c levels (HbA1c≥7.0) (Figure 5). Since HbA1c is an abnormally glycosylated hemoglobin (Bunn et al, 1976), which reflects overall blood glucose control (Koenig et al, 1976), the reductions of HbA1c might indicate that the long-term administration of WA improved the overall blood glucose control in diabetic patients with mild hyperglycemia. This was consistent with the suppression of postprandial hyperglycemia as shown in the single administration experiment.

The practical goals of diabetic treatment are to optimize glycemic control to prevent secondary complications primarily caused by hyperglycemia. In Japan, a prospective study in type II diabetic patients indicates that reduction of HbA1c levels less than 6.5% should be attempted for the prevention of microangiopathies (Ohkubo et al, 1995), with these criteria also being recommended for the prevention of macroangiopathies in type II diabetic patients. To achieve this, early combination therapy, including an insulin secretagogue, biguanide, insulin sensitizer, and/or insulin if HbA1c remains higher than 8%, is recommended (Rosenstock, 2000). However, although these approaches have been proven to decrease the risk of microangiopathies and macroangiopathies, they do increase the risk of hypoglycemia (UKPDS, 1998). The total incidence of hypoglycemia in type II diabetic patients using insulin was reported to be 36.5% (UKPDS, 1998) and 28% (Frier, 2002). Rates for hypoglycemic episodes caused by sulfonylureas were 11.0% for chlorpropamide and 17.7% for glibenclamide (UKPDS, 1998). Recently, a safer insulinotropic agent, nateglinide, has been introduced. It has not induced severe hypoglycemia, but still induced hypoglycemia in 26.7% of the subjects receiving 120 mg of nateglinide per day (Saloranta et al, 2002). Therefore, hypoglycemia is considered to be a serious problem in DM treatment (Frier, 2002).

Among available antidiabetic agents, those delaying glucose absorption can be used without the risk of hypoglycemia. Acarbose, miglitol, and voglibose are the α-glucosidase inhibitors utilized as pharmacological agents in this category. Josse et al (2003) conducted a randomized clinical trial to determine the effect of acarbose and observed no hypoglycemic events in the acarbose group (n=93). In a randomized control trial of miglitol and placebo, the incidence of hypoglycemia in the patients taking miglitol was 10% (25 mg TID) and 9% (50 mg TID), which was similar to that of the placebo group (8%). Thus, in fact, these agents showed less hypoglycemic events than other categorized agents. However, these agents often induce gastrointestinal symptoms. Josse et al (2003) reported that acarbose induced flatulence (80 vs 25% in the placebo group) and diarrhea (31 vs 9% in the placebo group). The recent STOP-NIDDM trial (Chiasson et al, 2002) confirmed that patients taking acarbose experienced flatulence (68 vs 27% in the placebo group) and diarrhea (32 vs 17% in the placebo group). Miglitol induced gastrointestinal symptoms such as flatulence (34%), diarrhea (23%), and nausea (11%) (Johnston et al, 1998). Voglibose also causes gastrointestinal disturbances (10%) (Goto et al, 1992). These common adverse effects in association with the use of α-glucosidase inhibitors were considered to result from its mechanism of action as undigested carbohydrate reaches the colon and generates gas by fermentation, which then causes gastrointestinal symptoms such as flatulence, flatus, crepitus, and diarrhea (Baba, 1994). Sometimes ileus has been reported during the treatments with acarbose and voglibose (Ohno, 1995), and acute hepatotoxicity has been observed with acarbose (Carrascosa, 1997; Fujimoto, 1998; Diaz-Gutierrez, 1998) or voglibose (Kawakami et al, 2001).

In both the single and long-term studies involving the administration of WA, no episodes of hypoglycemia were observed. WA acts as an α-amylase inhibitor preventing postprandial hyperglycemia, but because of its subsequent rapid digestion and loss of activity, it is unlikely to cause hypoglycemia. The absence of gastrointestinal adverse events was another feature of WA, as compared with the high incidences observed during treatment with the α-glucosidase inhibitors (Josse et al, 2003; Chiasson et al, 2002; Baba, 1994).

Whereas the α-glucosidase inhibitor delays disaccharide digestion and absorption, probably resulting in disaccharide accumulation in the intestine, which could elevate the osmolarity and then irritate the gastrointestinal tract, the amylase inhibitor does not result in the accumulation of disaccharides, and therefore would not cause the symptoms associated with an irritated gastrointestinal tract (Baba, 1994). In addition, the amylase inhibitors in WA are rapidly catabolized by proteinases, with this contributing to the avoidance of gastrointestinal symptoms. Furthermore, because WA consists of amino acids like the other proteins in food, it would be most unlikely to cause hepatic disorders such as those observed with the α-glucosidase inhibitors (Fujimoto et al, 1998; Kawakami et al, 2001).

There are possible disadvantages of WA treatment, however. Since such treatment will not reduce the total energy intake, weight gain may occur if patients do not manage proper appetite control. Additionally, wheat α-amylase inhibitors are major allergens in wheat flour (Amano et al, 1998), so patients with wheat allergy would be unable to use them.

In addition to controlling blood glucose levels and reducing the vascular complications in diabetic patients, any treatment which would reverse the ever-increasing incidence of type II diabetes in the world would be of value. Recently, lifestyle modification, including dietary restriction and exercise promotion, was reported to be the most effective therapy available to reduce the incidence of type II diabetes in subjects with IGT in a large randomized intervention study (Knowler et al, 2002). Metformin, a biguanide, was also shown to reduce the development of DM in the same study (Knowler et al, 2002). α-Glucosidase inhibitors may have a similar ability to prevent DM by inhibiting the postprandial increase in blood glucose. In fact, the STOP-NIDDM Group reported that the α-glucosidase inhibitor, acarbose, could be used either as an alternative or in addition to changes in lifestyle to prevent the development of type II diabetes in subjects with IGT (Chiasson et al, 2002) and also to reduce the risk of cardiovascular disease and hypertension in such patients (Chiasson et al, 2003).

Amylase inhibitors and α-glucosidase inhibitors may have the potential to prevent the development of diabetes in subjects with IGT, and amylase inhibitors may have an advantage over the α-glucosidase inhibitors because of the much lesser incidence of gastrointestinal side effects. WA is also expected to have this potential as an amylase inhibitor.

In conclusion, WA was demonstrated to suppress postprandial hyperglycemia, resulting in improved overall glycemic control in type II diabetic patients with mild hyperglycemia without the occurrence of critical adverse effects such as hypoglycemia, gastrointestinal symptoms, or hepatotoxicity. Consequently, WA may be a useful functional food, which could assist in the blood glucose control in type II diabetic patients with mild hyperglycemia with safety. Furthermore, WA may be useful for preventing diabetes in subjects with impaired glucose tolerance.

Additional information

Guarantor: S Inoue.

Contributors: This study was designed by TK and SI. SI, YN, JS and YT contributed to the subject recruiting and clinical data collection. TK, TM and IK carried out in vitro experiments and sample preparation. YS contributed to the manuscript preparation. SI was the main person responsible for all stages of the study.


  1. Amano M, Ogawa H, Kojima K, Kamidaira T, Suetsugu S, Yoshihama M, Satoh T, Samejima T & Matsumoto I (1998): Identification of the major allergens in wheat flour responsible for baker's asthma. Biochem. J. 330, 1229–1234.

  2. Baba S (1994): Alpha-glucosidase inhibitor. In Novel Development in Pharmacological Therapy of Diabetes. Baba S pp 53–54. Japan Churchill Livingstone. (in Japanese).

  3. Bunn HF, Haney DN, Kamin S, Gabbay KH & Gallop PM (1976): The biosynthesis of human hemoglobin A1c. J. Clin. Invest. 57, 1652–1659.

  4. Carrascosa M, Pascual F & Aresti S (1997): Acarbose-induced acute severe hepatotixicity. Lancet 349, 698–699.

  5. Chiasson JL, Josse RG, Gomis R, Hanefeld M, Karasik A, Laakso M & STOP-NIDDM Trial Research Group (2002): Acarbose for prevention of type 2 diabetes mellitus: the STOP-NIDDM randomized trial. Lancet 359, 2072–2077.

  6. Chiasson JL, Josse RG, Gomis R, Hanefeld M, Karasik A, Laakso M & STOP-NIDDM Trial Research Group (2003): Acarbose treatment and the risk of cardiovascular disease and hypertension in patients with impaired glucose tolerance: the STOP-NIDDM trial. J. Am. Med. Assoc. 290, 486–494.

  7. Choudhury A, Maeda K, Murayama R & DiMagno EP (1996): Character of a wheat amylase inhibitor preparation and effects on fasting human pancreaticobilary secretions and hormones. Gastroenterology 111, 1313–1320.

  8. DePonte R, Parlamenti R, Petrucci T, Silano V & Tomasi M (1976): Albumin α-amylase inhibitor families from wheat flour. Cereal Chem. 53, 805–820.

  9. The Diabetes Control and Complications Trial Research Group (1993): The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. New Engl. J. Med. 329, 977–986.

  10. Diaz-Gutierrez FL, Ladero JM & Diaz-Rubio M (1998): Acarbose-induced acute hepatitis. Am. J. Gastroenterol. 93, 481.

  11. Frier BM (2002): Hypoglycemic valleys: an under-recognised problem in type 2 diabetes? Int. J. Clin. Pract. Suppl. 129, 12–19.

  12. Fujimoto Y, Ohhira M, Miyokawa N, Kitamori S & Kohgo T (1998): Acarbose-induced hepatic injury. Lancet 351, 340.

  13. Goto Y, Baba S, Nakagawa S, Kuzuya T, Hirata Y, Akunuma Y, Sakamoto N, Shigeta Y & Mimura G (1992): The availability for non-insulin dependent diabetes mellitus of α-glucosidase inhibitor AO-128. Phase III comparison test between double blind groups. Igaku No Ayumi 160, 943–971. (in Japanese).

  14. Japanese Ministry of Health and Public Welfare (1999): National survey of diabetes mellitus. 1997 (in Japanese).

  15. Japanese Ministry of Health, Labor, and Public Welfare (2003): Brief report of the national survey of diabetes mellitus. 2002 (in Japanese).

  16. Johnston PS, Lebovitz HE, Coniff RF, Simonson DC, Raskin P & Munera CL (1998): Advantage of α-glucosidase inhibition as monotherapy in elderly type 2 diabetic patients. J. Clin. Endocrinol. Metab. 83, 1515–1522.

  17. Josse RG, Chiasson JL, Ryan EA, Lau DCW, Ross SA, Yale JF, Leiter LA, Maheux P, Tessier D, Wolever TMS, Gerstein H, Rodger NW, Dornan JM, Murphy LJ, Rabasa-Lhoret R & Meneilly GS (2003): Acarbose in the treatment of elderly patients with type 2 diabetes. Diabetes Res. Clin. Pract. 59, 37–42.

  18. Kashlan N & Richardson M (1981): The complete amino acid sequence of a major wheat protein inhibitor of α-amylase. Phytochemistry. 20, 1781–1784.

  19. Kawakami S, Arima T, Harada K, Miyazono H, Oketani M, Miyazaki H & Arima T (2001): Hepatic necrosis with cholestasis induced by long-term voglibose administration. Intern. Med. 40, 484–488.

  20. Kneen E & Sandstedt RM (1943): An amylase inhibitor from certain cereals. J. Am. Chem. Soc. 65, 1247.

  21. Knowler WC, Barrett-Connor E, Fowler SE, Hamman RF, Lachin JM, Walker EA, Nathan DM & Diabetes Prevention Program Research Group (2002): Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. New Engl. J. Med. 346, 393–403.

  22. Koenig RJ, Peterson CM, Jones RL, Saudek C, Lehrman M & Cerami A (1976): Correlation of glucose regulation and hemoglobin A1c in diabetes mellitus. New Engl. J. Med. 295, 417–420.

  23. Koike D, Yamadera K & DiMagno EP (1995): Effect of a wheat amylase inhibitor on canine carbohydrate digestion, gastrointestinal function, and pancreatic growth. Gastroenterology 108, 1221–1229.

  24. Maeda K, Hase T & Matsubara H (1983): Complete amino acid sequence of an alpha-amylase inhibitor in wheat kernel. Biochim. Biophys. Acta 743, 52–57.

  25. Oda Y, Matsunaga T, Fukuyama K, Miyazaki T & Morimoto T (1997): Tertiary and quaternary structures of 0.19 α-amylase inhibitor from wheat kernel determined by X-ray analysis at 2.06-A resolution. Biochemistry 36, 13503–13511.

  26. Ohkubo Y, Kishikawa H, Araki E, Miyata T, Isami S, Motoyoshi S, Kojima Y, Furuyoshi N & Shichiri M (1995): Intensive insulin therapy prevents the progression of diabetic microvascular complications in Japanese patients with non-insulin-dependent diabetes mellitus: a randomized prospective 6-year study. Diabetes Res. Clin. Pract. 28, 103–117.

  27. Ohno T (1995): Investigation of gastrointestinal adverse effects by alpha-glucosidase inhibitor, especially the development of ileus. J. New Remedies Clin. 44, 1114–1116. (in Japanese).

  28. Rosenstock J (2000): Treatment strategies and new therapeutic advances for type 2 diabetes. Diabet. Educ. 26 (Suppl), 14–18.

  29. Saloranta C, Guitard C, Pecher E, De Pablos-Velasco P, Lahti K, Brunel P & Groop L (2002): Nateglinide improves early insulin secretion and controls postprandial glucose excursions in a prediabetic population. Diabetes Care 25, 2141–2146.

  30. Scherbaum WA & Zimmet P (1995): The prevention of the NIDDM epidemic. IDF Bull. 40, 6–16.

  31. Sodini G, Silano V, de Agazio M, Pocchiari F, Tentori L & Vivaldi G (1970): Purification and properties of a Triticum aestivum specific albumin. Phytochemistry 9, 1167–1172.

  32. Stratton IM, Kohner EM, Aldington SJ, Turner RC, Holman RR, Manley SE & Matthews DR (2001): UKPDS 50: risk factors for incidence and progression of retinopathy in type II diabetes over 6 years from diagnosis. Diabetologia 44, 156–163.

  33. UK Prospective Diabetes Study (UKPDS) Group (1998): Intensive blood-glucose control with sulpfonylureas or insulin compaired with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet 352, 837–853.

  34. Yanai H (2004): Statcel—The Usefull Add-in Software Forms on Excel 2nd Edition, Tokyo, Japan. OMS (in Japanese).

Download references

Author information

Correspondence to S Inoue.

Rights and permissions

Reprints and Permissions

About this article


  • wheat albumin
  • α-amylase inhibitor
  • diabetes mellitus
  • postprandial blood glucose
  • hemoglobin A1c (HbA1c)