Nutrition and metabolism (Module 4)

Overview on aims and the objectives in the 5th project year (June 2009 - May 2010)

Observational data, including results from a limited number of interventions, have indicated a protective role of a whole grain diet adjunct to development of the metabolic syndrome. However, the crucial cereal factor for these benefits is not known, and some recent interventions trials have found but modest benefits on metabolic risk parameters. The work packages in the Nutrition and metabolism module has been organized to address this specific topic, focusing on the role of certain bioactive components present in whole grain products, including the dietary fibre components and other indigestible carbohydrate fractions; and on the role of the glycaemic index (GI) characteristics. Major focus has been on wheat and to some extent rye, representing the major bread cereals in Europe. However, some mechanistic aspects have also been studied using barley as ingredient material. The bioavailability of various bioactive components in whole grain has been evaluated in vitro, in animal experiments and in meal studies in humans. The animal experimental data have been reported in full previously. For certain whole grain components, the importance of the colonic microflora in mediating the metabolic effects has been evaluated. A mechanistic approach was also intended when designing the longer-term intervention studies, aiming at addressing effects on acknowledged risk factors for type 2 diabetes, cardiovascular disease and obesity in relation to the content and bioavailability of specific whole grain components and the potential additive value of whole grain products which also are low in GI and/or Insulinaemic Index (II).

The objectives for the reporting period were

• To evaluate the bioavailability of certain bioactive components, and the metabolic effects of such components and other whole grain product characteristics using acute and semi-acute meal studies in healthy subjects. A special focus with respect to the acute meal studies were to investigate the potential of different genotypes from rye and wheat on glycaemic and insulinaemic responses.
• To execute and/or evaluate the longer-term interventions in at risk subjects

Work performed and the main achievements

WP4.1 Product characteristics, release features and bioavailability of components associated with the ”whole grain concept” – studies using in vitro models simulating the human gastrointestinal tract

Previous work within this WP employing in vitro models indicative of small intestinal bioavailability (TIM-1) and colonic fermentability and metabolism (TIM-2), respectively, reported that the aleurone fraction of wheat contained potentially bioactive components, and that these components were metabolised by human gut microbiota in vitro. Relative to other analysed compounds in wheat fractions (aleurone, bran, flour), ferulic acid was the most important bioactive compound. The anti-oxidant capacity of wheat fractions was highly correlated with the amount of aleurone in the fraction. Ferulic acid was responsible for 60% of the antioxidant capacity of wheat aleurone. It was further shown that the bio-accessibility of ferulic acid from aleurone and bran, as well as from aleuroneenriched bread, was low (1-2%) in a simulated upper-gut in vitro model (TIM-1). Innovative processing and pretreatment of bran (lactic acid fermentation, enzymes) increased the bioaccessibility of ferulic acid and other phenolic acids (coumaric- and sinapic acid) in the TIM-1 model. The bioaccessible fraction of ferulic acid was essentially in free form. Results from experiments of the anti-inflammatory potential of certain compounds in a cell bioassay have been achieved. The bio-accessible fraction collected from aleurone in the TIM-1 model showed prolonged anti-inflammatory potential as judged from prolonged inhibition of TNF-α in a LPS induced human macrophage cell system.

Most of the phenolic compunds passed the small intestine, and entered the colon under simulated in vitro conditions (TIM-2 model), where they can be released and further metabolized by the colonic microbiota. Metabolites formed from phenolic compounds in the TIM-2 model have been identified e.g. phenyl propionic acid derivatives; with possible health benefits. Studies have further shown that pre-treatment of bran (lactic acid fermentation and/or enzyme treatment) increased butyric acid formation from a wheat bran bread in the TIM-2 model, compared with white bread, whole grain wheat bread or bread with native bran. This metabolite originating from microbial metabolism may also be associated with health benefits.

As a follow up to the above results, an in vivo study has been conducted during the present reporting period in order to confirm the above in vitro results. The in vivo study was performed as an acute meal study in healthy volunteers and focused on the importance of processing of wheat bran for the bioavailability of ferulic acid and related phenolics, the antioxidant capacity, and the inflammatory response. The wheat bran was processed employing yeast fermentation and enzyme treatment for 20h at 20°C, and incorporated into a bread (test bread). A non-treated wheat bran was incorporated into a control bread. The bran fractions were included at 9% level. The meal study was performed using a randomized cross-over design, and was preceeded by a 3 d period with a diet low in phenolics. Eight healthy male subjects (BMI<30) participated. Bioprocessing of wheat bran led to a 2 to 3- fold increase in the bioavailability of ferulic acid, vanillic and sinapic acids and 3,4- dimethoxy benzoic acidin the test bread product.There were no differences in total anti-oxidant capacity in plasma following the test bread or control bread, respectively. Prior to and after the bread meals, blood was withdrawn from the subjects, and subsequently a test of inflammatory response was studied ex-vivo. The potential anti-inflammatory effect was assessed by measurement of the ratio of pro-inflammatory vs. anti-inflammatory cytokines in a cell bio-assay using LPS to promote inflammation. It was found that plasma derived from subjects provided the test bread led to a decreased proportion of pro-inflammatory cytokines ex-vivo, suggesting that bio-processing may enhance antiinflammatory potential of whole grain wheat products.

WP 4.3 Bioavailability and acute metabolic effects of components associated with the “whole grain concept”- acute and semi-acute meal studies in healthy and “at risk” subjects

From two series of acute meal studies regarding the bioavailability of methyl donors and ferulic acid in healthy subjects it was concluded that ingestion of raw bran and aleurone meals, showed significant post-meal increases in plasma betaine (two-fold increase from baseline) and plasma ferulic acid (five–fold increase from baseline). A second series of experiments with aleurone rich bread rolls confirmed the above results and demonstrated significant post-meal increases in plasma ferulic acid (three-fold increase from baseline) and betaine (two-fold increase from baseline). Adding iron to bread did not affect ferulic acid uptake. Effects on plasma choline and folate were small and non-significant, and no significant effects were observed in either of the meal studies on plasma tocopherols and carotenoids. The studies provide the first evidence of wheat products impacting massively on betaine status, and also novel evidence of wheat products impacting favourably on plasma ferulic acid. During the present reporting period, data from a long term intervention with an aleurone rich diet vs a control diet have been compiled and complementary analyses of biomarkers performed (se WP 4.4).

Previous studies in this WP indicated improved insulin economy with rye as compared with wheat products. Interestingly, the improved insulin economy with rye products was observed also with rye products made from endosperm rye. It was concluded that kernel based products, and rye bread with lactic acid displayed lower GI than white wheat bread (WWB). All rye breakfast meals, irrespectively of extraction rate or processing method, and the wheat kernel breakfast displayed lower II than WWB. The rye kernel breakfast produced a glucose increment of longer duration, but of a lower magnitude than WWB and wheat kernels. Moreover, the rye kernel breakfast induced a higher satiety, promoted a lower desire to eat; and also reduced voluntary intake at a subsequent ad libitum lunch by 15%. During the previous reporting period a comparison of glucose and insulin responses to five different rye genotypes has been evaluated revealing that not all rye genotypes have insulin saving properties, as previously thought when using rye blends. The rye genotypes were obtained from Module 2 and included; Haute- Loire Pop, Nikita, Dankowski Zlote, Rekrut and Amilo. The results indicated differences in metabolic responses when served to healthy subjects in carbohydrate equivalent meals. Consequently, only Rekrut and Amilo displayed lowered insulinaemic responses compared to a white wheat reference bread. Additionally correlations were made more recently between contents of potentially bioactive components and insulinaemic responses. The results indicate that total alkylresorcinols, folate and phenolic acids were correlated with beneficial effects on glucose and/or insulin. Increased levels of total tocols was associated with increased glucose and insulin responses, and sterols varied in their effect on glucose and insulin.

Meal studies have also been performed in healthy subjects using a unique high-amylose wheat variety obtained from Module 2. The flour was included in bread and baked at low temperature, long-time baking condition in order to promote the formation of slowly digestible starch and enzyme resistant starch (RS). Bread baked with highamylose wheat genotype contained significantly higher amounts of RS (total starch basis) than similar bread baked with commercial whole grain wheat flour and white wheat reference bread, respectively. A significant reduction in GI was observed in breads baked from high-amylose genotype compared to the reference bread. It was concluded that high-amylose wheat show interesting properties and may be preferable to other wheat genotypes considering RS formation and glycaemic response.

Further, during the current reporting period, a meal study has also been conducted and evaluated, addressing the potential prebiotic effect of arabinoxylan-oligosaccharides (AXOS) derived from wheat. The study was performed in 20 healthy volunteers (10M;10W). The test products were served as evening meals and were as follows; 1. High GI white wheat bread (WWB), 2. WWB with AXOS included (12% dry basis) and RS (High-maize 260, 11% dry basis), 3. WWB with AXOS included (23% dry basis), and 4. WWB with RS (22% dry basis). At a subsequent standardised WWB breakfast fasted and post-prandial blood was collected for analysis of glucose tolerance, insulin and related markers (e.g. FFA). Additionally, measurement of subjective appetite rating and of markers of colonic fermentation (breath hydrogen) were performed. Fasting H2 and mean H2 (0-180 min), respectively, were positively associated with satiety. Low glucose responses at the standardized breakfast were associated with low levels of fasting free fatty acids (FFA) in the morning. Also, low postprandial glucose responses at breakfast was related to higher satiety ratings. The present study confirms previous results were overnight colonic fermentation of prebiotic carbohydrates was associated with improved glucose tolerance, suppressed FFA-levels and increased satiety. It is concluded that AXOS shows interesting potential as a prebiotic substrate with potential beneficial influence on glucose tolerance and appetite.

WP 4.4 Long-term effects of whole grain and associated product features on metabolic risk factors and weight regulation – studies in “at risk” subjects and in type 2 diabetics

Four longer-term interventions have been executed over the course of the project. One study focusing on the impact of whole grain wheat vs. refined wheat on weight loss in overweight women was initiated in May 2007 and completed in December 2007, with weight loss as primary end-point. It could be concluded that both hypo-caloric diets were effective in inducing weight loss, but whole grain foods had a more beneficial effect on body fat percentage and cholesterol levels than refined grains. It should be noted, however, that the whole grain diet tended to induce a more prominent sloop in weight loss over-time, with more accentuated differences appearing in the later intervention period. This may suggest that the 12w intervention period selected was too short to reveal potential benefits of whole grain wheat in this context. This is note-worthy, and it is of interest to study further the potential of whole grain diets employing a longer intervention period.

A 4w intervention trial in over-weight subjects with an aleurone rich test diet vs. a refined high-fibre control diet was completed in 2009, and analysis and evaluation has been performed during the final reporting period. The intervention was performed in 80 healthy subjects 45-70 yr with BMI >25. The main focus of the study was to exploit whether aleurone rich products, selected on basis of their potential to increase plasma levels of betaine and ferulic acid, positively affected blood lipids and other metabolic parameters. Results show that compared to a highfibre control the longer-term consumption of aleurone-rich products leads to; i) a significant decrease in plasma CRP but no change in other markers of inflammation, endothelial function or antioxidant activity, ii) a significant increase in plasma betaine and an associated decrease in homocysteine, with no change in status of other methyl donor micronutrients, iii) a significant decrease in plasma LDL-cholesterol but no change in other lipid profile measures, iv) significant increase in butyric acid levels, v) a significant increase in plasma insulin but no change in glucose levels. These findings provide potential mechanisms whereby whole grain may be beneficial to health. During the last reporting period additional analyses of SCFA have been performed in collaboration with Aarhus University. Results indicate similar fasting plasma total SCFA levels, but with an increased proportion of butyric acid following the aleurone rich diet.

During the final year the intervention study in diet-treated type 2 diabetic patients, was completed and evaluated. The study was performed over 28d using 4 parallel and matched groups and included 60 subjects evenly distributed over the 4 arms. The diets varied with respect to the GI features and/or whole grain content of the bread products. The control diet included high GI/white wheat bread; whereas the three different test diets contained either high GI/whole grain wheat flour bread, low GI/whole grain wheat kernel bread, or low GI/whole grain barley kernel bread, respectively. Insulin first-phase response and insulin sensitivity was used as critical end points employing a recently developed procedure, the Glucagon Insulin Tolerance Test (GITT-test), which is considered more informative than conventional clamp techniques. With GITT measurements of both insulin secretion and insulin sensitivity are combined. Additionally HbA1c, fasting plasma glucose, HDL, LDL, TG, ApoA, PAI-1 and high sensitivity CRP were measured as secondary end-points. No differences were noted in insulin sensitivity from day 1 to 28 following the intervention with high GI white bread diet, high GI whole grain wheat flour bread diet, or low GI whole grain wheat kernel bread diet. In contrast, in the case of the low GI whole grain barley diet there was a statistically significant increase in insulin sensitivity day 28 compared to day 1 as measured by GITT (k-value 2.01 compared with 1.44, P<0.001). There were no statistically significant differences in any of the secondary endpoints with any of the bread diets. It was concluded that enclosure of low GI whole grain barley kernel bread in the diet improve insulin sensitivity over 4 w in a group of patients with diet-treated type 2 diabetes.

Previous work within WP4.3 has demonstrated that low GI barley kernel bread, when ingested in the evening, improve glucose tolerance the subsequent breakfast in healthy subjects. The effect could be assigned to colonic fermentation of the dietary fibre and resistant starch present in this type of whole grain bread product, and was associated with increased levels of plasma GLP-1 during the breakfast meal. It is put forward that the benefits seen on insulin sensitivity in type 2 diabetics following the intervention with low GI barley kernel bread (reported above) might be related to the release of the incretin GLP-1 with acknowledged anti-diabetic properties.

A joint two-centre study was performed to determine the effects of a diet with multiple beneficial characteristics of whole grains on glucose and insulin metabolism in subjects with metabolic syndrome in two different dietary backgrounds (Kuopio/Finland and Naples/Italy). Subjects fulfilling the NCEP criteria for metabolic syndrome were included; with about 30 in each treatment arm, using a parallel design. The intervention period was 12 weeks. In Kuopio, whole grain rye and sour-dough rye bread were included in the test diet to secure whole grain and low GI/and or II features. Whole grain pasta and barley soup are examples of products included in the test diet in Naples. Compared with base-line, no changes were seen in anthropometric parameters (body weight, BMI, waist circumference) over the 12w period. Similarly, there were no changes in fasting plasma glucose, serum insulin, or blood lipids during the intervention period for any of the treatment arms. FSIVGTT (frequently sampled intravenous glucose tolerance test) did not reveal any effect on insulin sensitivity due to treatment for any of the diets. However, an OGTT indicated a non significant tendency towards improved glucose tolerance following the whole grain diet in the Kuopio setting. In the Naples study, there was a reduced post-prandial need for insulin following a meal representative of the whole grain intervention period. Additionally, post-prandial triglyceride clearance was improved by provision of a meal representative of the whole grain intervention period as opposed to a meal representing the refined control diet.

Over-all conclusions

Meal studies in healthy subjects indicated metabolic benefits of certain barley and rye products. Also, aleurone rich products promoted higher plasma levels of potentially bioactive components such as betaine and ferulic acid.

As judged from the semi-long term interventions, whereas ordinary whole grain wheat appeared not to be effective in reducing metabolic risk factors or facilitating weight loss (over-weight women), enclosure of whole grain wheat products enriched with wheat aleurone, significantly reduced serum LDL-cholesterol and inflammatory response (CRP) (healthy over-weight)

Low GI whole grain barley kernel products improved glucose tolerance and satiety in the perspective from an evening test meal to a standardised breakfast in healthy subjects. The benefits were related to increased GLP-1 levels at the time of the breakfast, possibly mediated by colonic fermentation of the dietary fibre and RS present in this type of products. Enrichment of white wheat bread with AXOS also tended to improve glucose tolerance and satiety in the perspective from evening meal to a standardised breakfast. The benefits of low GI whole grain barley kernel bread on glucose metabolism was confirmed in diet treated type 2 diabetics, as judged from improved insulin sensitivity following a 4w intervention with this type of bread.

It is concluded that whole grain diets may improve insulin sensitivity and metabolic risk factors through mechanisms related to increased colonic fermentation of indigestible carbohydrates and stimulation of GLP-1. Also, whole grain diets may lower plasma cholesterol through a mechanism related to lowered homocystein levels, mediated through the provision of the methyl-donor betaine.