Centrally administered glucagon-like peptide-1 (GLP-1) inhibits feeding in fasted rats, but its role in human satiety has been largely unexplored. The present study investigated the effect of peripheral GLP-1 infusion on gastric emptying and satiety in man. Ten non-obese male subjects were infused in a randomized single-blind within-subject crossover study using saline infusion as control. They received either a GLP-1 infusion (1·2 pmol/kg per min) or a saline infusion for 1 h, at 18.00 hours. At 20 min after starting the infusion the gastric emptying of a 400 ml water load was measured. Subjects completed behavioural self-rating scales to assess hunger and satiety. After 40 min subjects were given a buffet meal ad libitum and their food intake was recorded. GLP-1 infusion raised circulating GLP-1 concentrations to approximately twice those seen following a meal. It did not affect circulating insulin levels but caused a small fall in glucose levels. Gastric emptying of the water load was significantly delayed by the GLP-1 infusion. Energy intake from the buffet was unaffected by GLP-1 infusion. Self-assessment of hunger and satiety was similarly unaffected by the infusion before the buffet meal, although subjects tended to be less hungry after the buffet meal following GLP-1 infusion (P < 0·09). GLP-1 infusion delayed gastric emptying but had a minimal effect on food intake and satiety. This study casts doubts on whether GLP-1 is a major satiety factor in man, although a raised circulating plasma glucose level, as would normally occur postprandially, might be necessary for GLP-1 to increase satiety.

Lime-treated maize husks (LTCH), a by-product of tortilla manufacturing in Mexico, have been shown to decrease plasma LDL-cholesterol levels in guinea-pigs by specific alterations in hepatic cholesterol metabolism. To determine whether LTCH would also lower plasma cholesterol levels in normal and hyperlipidaemic individuals, the fibre content of a typical diet was increased by supplementing free-living subjects in North-West Mexico with cookies containing 450 g LTCH/kg. Normal subjects (n 11) with plasma cholesterol levels of less than 5·7 mmol/l and hypercholesterolaemic subjects (n 12) with plasma cholesterol levels higher than 5·7 mmol/l participated in the study. Plasma glucose, cholesterol, triacylglycerol, LDL-cholesterol and HDL-cholesterol concentrations, LDL: HDL values and blood pressure were determined at baseline and after 6 weeks of supplementation with LTCH. LTCH supplementation significantly lowered the plasma total cholesterol level by 11–15 % and LDL-cholesterol by 25 %, and improved the LDL: HDL value by 29–33 % (P < 0·01) in both normal and hypercholesterolaemic subjects. HDL-cholesterol, triacylglycerol and glucose concentrations did not change. Both groups consumed equal amounts of LTCH per week; individuals showed excellent compliance and good acceptance of the product. Neither group changed their dietary habits during the time of the experiment as determined by 3 d dietary records at baseline and at week 6. We conclude that LTCH supplements are suitable to increase fibre intake and reduce plasma LDL-cholesterol levels in healthy and hypercholesterolaemic subjects in this population.

The effects of different polyphenol-containing beverages on Fe absorption from a bread meal were estimated in adult human subjects from the erythrocyte incorporation of radio-Fe. The test beverages contained different polyphenol structures and were rich in either phenolic acids (chlorogenic acid in coffee), monomeric flavonoids (herb teas, camomile (Matricaria recutita L.)), vervain (Verbena officinalis L.), lime flower (Tilia cordata Mill.), pennyroyal (Mentha pulegium L.) and peppermint (Mentha piperita L.), or complex polyphenol polymerization products (black tea and cocoa). All beverages were potent inhibitors of Fe absorption and reduced absorption in a dose-dependent fashion depending on the content of total polyphenols. Compared with a water control meal, beverages containing 20–50 mg total polyphenols/serving reduced Fe absorption from the bread meal by 50–70 %, whereas beverages containing 100–400 mg total polyphenols/serving reduced Fe absorption by 60–90 %. Inhibition by black tea was 79–94 %, peppermint tea 84 %, pennyroyal 73 %, cocoa 71 %, vervain 59 %, lime flower 52 % and camomile 47 %. At an identical concentration of total polyphenols, black tea was more inhibitory than cocoa, and more inhibitory than herb teas camomile, vervain, lime flower and pennyroyal, but was of equal inhibition to peppermint tea. Adding milk to coffee and tea had little or no influence on their inhibitory nature. Our findings demonstrate that herb teas, as well as black tea, coffee and cocoa can be potent inhibitors of Fe absorption. This property should be considered when giving dietary advice in relation to Fe nutrition.

Whole-body and gastrointestinal tract (GIT) metabolism of [5-15N]glutamine were monitored in lambs (33 kg live weight) fasted for 24 h. Animals were previously prepared with vascular catheters across the mesenteric-(MDV) and portal-drained viscera (PDV) to permit quantification of mass and isotopic transfers of metabolites by arterio-venous difference. Continuous infusions of [5-15N]glutamine into the jugular vein were conducted for 10 h and integrated blood samples withdrawn over 75 min intervals for the last 5 h of infusion. The lambs were then killed and portions from various tissues of the digestive tract and other body organs removed for determination of 15N enrichment in RNA, DNA and protein (the latter obtained by difference using total acid-precipitable N). Whole-body glutamine flux was 108 μmol/min of which 23 and 47 % could be attributed to MDV and PDV metabolism (P < 0·001) respectively. There was a small net production of glutamine across the MDV. GIT blood-flows and NH3 production were partitioned 3:2 between MDV and non-MDV components. Less than 5 % of the NH3 produced was derived from the amido-N of glutamine, while across the small intestine (MDV) 26 % of the glutamine flux was converted to NH3, compared with 18 % for non-MDV transfers. The 15N enrichments in protein were of the order jejunum > duodenum > ileum with mucosal cells more labelled than serosal (P < 0·001). Lesser enrichments were observed for other GIT tissues (abomasum > caecum > rumen) while liver and lymph were comparable with the abomasum; kidney, spleen and muscle were lower still (P < 0·05). Enrichments of RNA were similar to that of protein and followed the same pattern, except for higher relative values for liver, spleen and lymphoid tissue. The lowest enrichments were observed for DNA, but again the pattern order was similar except for increased label in lymph, caecum and the spleen. For the MDV there was reasonable agreement between 15N-disappearance as glutamine and appearance in NH3 (24 %), protein (81 %), RNA (3·6 %) and DNA (2·1 %). For the total PDV there was a shortfall (−12 %), however, which may be due to losses in lumen components. These results show the importance of the GIT as a contributor to total glutamine plasma flux, but indicate a lesser reliance on glutamine metabolism by the digestive tract of the ruminant compared with observations from non-ruminants.

The influence of different N sources on fermentation rate and de novo amino acid synthesis by rumen micro-organisms was investigated in vitro using rumen fluid taken from four sheep receiving a mixed diet comprising (g/kg DM): grass hay 500, barley 299·5, molasses 100, fish meal 91, minerals and vitamins 9·5. Pancreatic casein hydrolysate (P; comprising mainly peptides with some free amino acids; 10 g/l), free amino acids (AA; casein acid hydrolysate + added cysteine and tryptophan; 10 g/l), or a mixture of L-proline, glycine, L-valine and L-threonine (M; 0·83 g/l each) were added to diluted (1:3, v/v), strained rumen fluid along with 15NH4Cl (A; 1·33 g/l) and 6·7 g/l of a mixture of starch, cellobiose and xylose (1:1:1, by weight). P and AA, but not M, stimulated net gas production after 4 and 8 h incubation (P < 0·05) in comparison with A alone. P increased microbial-protein synthesis (P < 0·05) compared with the other treatments. All of the microbial-N formed after 10 h was synthesized de novo from 15NH3 in treatment A, and the addition of pre-formed amino acids decreased the proportion to 0·37, 0·55, and 0·86 for P, AA, and M respectively. De novo synthesis of amino acids (0·29, 0·42 and 0·69 respectively) was lower than cell-N. Enrichment of alanine, glutamate and aspartate was slightly higher than that of other amino acids, while enrichment in proline was much lower, such that 0·83–0·95 of all proline incorporated into particulate matter was derived from pre-formed proline. Glycine, methionine, lysine, valine and threonine tended to be less enriched than other amino acids. The form in which the amino acids were supplied, as P or AA, had little influence on the pattern of de novo synthesis. When the concentration of peptides was decreased, the proportion of microbial-N formed from NH3 increased, so that at an initial concentration of 1 g peptides/l, similar to the highest reported ruminal peptide concentrations, 0·68 of cell-N was formed from NH3. Decreasing the NH3 concentration at 1·0 g peptides/l caused proportionate decreases in the fraction of cell-N derived from NH3, from 0·81 at 0·53 g NH3-N/l to 0·40 at 0·19 g NH3-N/l. It was concluded that different individual amino acids are synthesized de novo to different extents by mixed rumen micro-organisms when pre-formed amino acids are present, and that the source of N used for synthesis of cell-N and amino acids depends on the respective concentrations of the different N sources available; however, supplementing only with amino acids whose synthesis is lowest when pre-formed amino acids are present does not stimulate fermentation or microbial growth.

We examined the effects of feeding guar-gum hydrolysate (GGH), a highly fermentable form of dietary fibre with low viscosity, on Ca absorption in the small and large intestines in rats under conditions in which gastric acid secretion was suppressed by a proton pump inhibitor, omeprazole. We also examined the role of the caecum in influencing these effects. The study was designed in a 2×2×2 factorial arrangement with two diet (GGH-containing (50 g/kg diet) and GGH-free diets) groups, two injection (omeprazole and vehicle) groups and two operation (sham and caecectomy) groups. Apparent Ca absorption was lower in rats administered omeprazole (30 mg/kg body weight per d) for 8 d than in rats administered the vehicle. Ingestion of GGH led to partial restoration of Ca absorption decreased by omeprazole treatment. However, this increment in Ca absorption was not sufficient to meet requirements because the dietary Ca level (3·0 g/kg diet) was the minimum requirement for the intact rats. The small increment in Ca absorption caused by the GGH diet was completely abolished by caecectomy. Soluble Ca pools in the caecal and colonic contents were increased by feeding GGH, and the soluble Ca concentrations were much higher than the Kt values of the Ca active transport system in the large intestine or the serum Ca concentration. These findings suggest that Ca solubilization is not a limiting factor for Ca absorption in the large intestine. Apparent Mg absorption was clearly lower in caecectomized rats than in sham-operated rats, and higher in the GGH-fed groups than in the groups fed on the GGH-free diet, even in the case of caecectomized rats. We conclude that Ca absorption lowered by inhibition of gastric acid secretion is partially restored in rats fed with GGH, but the increment is not sufficient to meet requirements.

The rates of utilization and oxidation of glutamine and glucose by oesophageal and duodenal tissues have been investigated in both rats and human subjects. In the rat, glutamine utilization by oesophageal tissue was 2–3-fold lower than that in the duodenum, and this substrate contributed less than 10 % to the total oxidative metabolism of the tissue, even when glutamine was the only substrate provided. In contrast, rat duodenal tissue derived about 34 % of the total CO2 production from glutamine-C, and this contribution was not suppressed by the addition of either glucose or a mixture of the other substrates. Rates of glucose utilization and oxidation by the duodenum were lower than those for glutamine, and were significantly (P<0·001) suppressed by addition of glutamine. In both oesophageal and duodenal tissues, less than 10 % of the glutamine-C utilized was fully oxidized, approximately 60–70 % was converted to glutamate, and 30–40 % to alanine. Results obtained using human biopsy tissue samples were similar to those observed in the rat. Glutamine oxidation contributed 34 (sd 4) % of the total CO2 production by the duodenal tissue, but only 8 (sd 4) % to oesophageal tissue oxidation. The findings suggest that glutamine is not an important or preferred fuel for oesophageal tissue, whereas it is for duodenal tissue. Thus, these tissues can be expected to respond differently to glutamine administration.

To investigate the influence of different dietary fats on lipids and anti-cardiolipin antibody levels, autoimmune NZB/W F1 mice were fed on diets containing 200 g dietary fat as palm oil, lard–soyabean oil (1:1, w/w), soyabean oil, rapeseed oil or fish oil/kg. In addition, each dietary fat group was divided into an early-feeding group with feeding from 2 months of age, and a late-feeding group with feeding from 5 months of age. Serum levels of triacylglycerol, phospholipid, cholesterol and anti-cardiolipin antibody were measured at regular intervals, and mice were killed at the age of 7 months for analysis of hepatic lipid and fatty acids. The results showed that hepatic triacylglycerol and cholesterol contents were lower in mice fed on fish oil than in those fed on palm oil. In contrast, hepatic phospholipid content was higher in mice of the fish oil group than in those of the other four dietary fat groups. Composition profiles for both hepatic and renal oleic acid (18:1n-9), linoleic acid (18:2n-6) and eicosapentaenoic acid (20:5n-3) were similar to those of the dietary fats in mice of both early-feeding and late-feeding groups. Fish oil intake decreased arachidonic acid (20:4n-6) concentration in kidney tissue but not in liver tissue. Serum triacylglycerol, cholesterol and phospholipid levels were lower in mice fed on fish oil than in those fed on palm oil. Immunoglobulin (Ig) M anti-cardiolipin antibody was lower for the fish oil group than for the other groups. The IgG anti-cardiolipin antibody level was significantly lower in mice fed on fish oil compared with that of the palm oil group only in the early-feeding group. There was a positive correlation between serum IgM anti-cardiolipin antibody and phospholipid levels (early-feeding group r 0·902, P < 0·05; late-feeding group r 0·894, P < 0·05). These findings suggest dietary fish oil may affect both lipid levels and anti-cardiolipin antibody, contributing to alleviation of the autoimmune process in autoimmune-prone NZB×NZW F1 mice.